{
  "name": "LLM Bottleneck open compatibility dataset",
  "version": "2026-10-03-13163f6c89cd",
  "snapshot": {
    "id": "ZOAGzSDaZvVT",
    "engineVersion": "2026-10-04.1",
    "scenarioVersion": 1,
    "modelCatalogue": {
      "version": 1,
      "generatedAt": "2026-10-03T21:32:36.346Z",
      "models": 327
    },
    "hardwareCatalogue": {
      "version": 2,
      "generatedAt": "2026-09-24T13:48:13.988Z",
      "devices": 135
    },
    "inputs": [
      {
        "file": "data/model-catalog.json",
        "sha256": "626019e6c4856bef4521a3589a52e8769bb23ed0235d469112eb42909a39a1e3"
      },
      {
        "file": "data/hardware-catalog.json",
        "sha256": "4c7e772119275c104dc0183b8cc5483e5fe348cbdffc58cbe9cf233383074b11"
      },
      {
        "file": "data/weight-catalog.json",
        "sha256": "17630367ce04a3142f0d70eec8cc4fc5abaf14cd1a6630caf55b2e2f395cd201"
      },
      {
        "file": "data/gguf-catalog.json",
        "sha256": "9dfc099aa8b8e75001e2b9993314135d90fdd05615849902c7f6524755ee9c1c"
      },
      {
        "file": "data/gguf-tensor-catalog.json",
        "sha256": "9adc2da7e9663108f6839774b998bdd870cc26452cfcd431029b9193a2e24aab"
      },
      {
        "file": "data/model-documentation-facts.json",
        "sha256": "4f95b47f06cfce81c53800c72414a9903607681979807f4f22bb879d4da0620e"
      },
      {
        "file": "data/model-parameter-facts.json",
        "sha256": "9c89c7463af0257298d5568a7576dfe5ca40976a84badee4b79941a259e0e2a1"
      }
    ]
  },
  "context_tokens": 8192,
  "license": "CC BY 4.0",
  "licenseUrl": "https://creativecommons.org/licenses/by/4.0/",
  "licenseScope": "Applies to the compilation, the computed columns and the evidence labelling only. Model weights, model cards, configurations, product names, trademarks and manufacturers' specifications belong to their owners and are not relicensed; they are cited by source URL.",
  "attribution": "LLM Bottleneck open compatibility dataset (llmbottleneck.com/data), version 2026-10-03-13163f6c89cd, CC BY 4.0.",
  "counts": {
    "rows": 44145,
    "rowsByStatus": {
      "not_sizable": 945,
      "ok": 40095,
      "context_exceeds_model_max": 3105
    },
    "models": 327,
    "devices": 135,
    "modelsWithAnalysedRows": 297,
    "modelsNotSizable": 7,
    "modelsBelowReferenceContext": 23,
    "fittingRows": 23991
  },
  "fields": {
    "model_slug": "Catalogue identifier of the model, and the last path segment of its page.",
    "model_name": "Name as its publisher writes it.",
    "publisher": "Namespace the configuration was read from (for a mirror, the mirror's namespace).",
    "hf_repo": "Hugging Face repository the architecture was read from.",
    "config_revision": "Immutable commit of that repository, so the row can be reproduced.",
    "parameters": "Parameter count from the published safetensors metadata; empty when unusable.",
    "device_id": "Catalogue identifier of the device, and the last path segment of its page.",
    "device_name": "Product name as its manufacturer writes it.",
    "vendor": "Manufacturer of the device: NVIDIA, AMD, Apple or Intel.",
    "device_memory_gb": "Memory the manufacturer publishes, counted as GB × 10⁹ bytes. For a device sold in several configurations, the largest.",
    "device_bandwidth_gb_s": "Published memory bandwidth of that configuration.",
    "bandwidth_evidence": "verified when the manufacturer publishes the bandwidth; derived when it is arithmetic over a published bus width and data rate.",
    "context_tokens": "Context every row is sized at (8192).",
    "status": "ok: analysed. context_exceeds_model_max: the model's published maximum is below the reference context. not_sizable: no usable size source for this model. unsupported: the engine refused, see status_reason.",
    "status_reason": "Why a row is not ok; empty when ok.",
    "fits": "For status ok: true when at least one format fits with the whole model resident and no offload. Empty otherwise.",
    "best_format": "Highest-quality format that fits, empty when none does.",
    "required_bytes": "Memory that format needs: weights, KV cache and runtime reserve.",
    "utilisation": "required_bytes divided by device memory, at the best format that fits, or at the smallest format when none does.",
    "weight_size_evidence": "verified: a published weight file; derived: reconstructed from the architecture; bounded: a range from the parameter count.",
    "smallest_format": "Smallest format the model ships in, which is what a 'no' is measured against.",
    "shortfall_bytes": "How far the smallest format misses by, empty when something fits."
  },
  "rows": [
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-1-audio-10b-a1-8b",
      "model_name": "GigaChat3.1-Audio-10B-A1.8B",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
      "config_revision": "bf73d03a43bdf5118f5a4dbdc24ba6f56ac31cfb",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 174509750396,
      "utilisation": 0.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 255938811560,
      "utilisation": 0.9998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 269957533589,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 269957533589,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 360805423617,
      "utilisation": 0.8352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 3.6356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 8.7255,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.0907,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.7267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.8178,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 174509750396,
      "utilisation": 0.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 4.8475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 466876769598,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.7267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.7267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 466876769598,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 17.451,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.1814,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.1814,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 174509750396,
      "utilisation": 0.9695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 269957533589,
      "utilisation": 0.9998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.3634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 15.8645,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 43.6274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 170509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 15.8645,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 43.6274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 170509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 17.451,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 29.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 21.8137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 14.5425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 10.9069,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 5.4534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.1814,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.1814,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.2377,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.2377,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 7.2712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 3.6356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 3.6356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 3.6356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 3.6356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 2.4237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 174509750396,
      "utilisation": 1.8178,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78509750396
    },
    {
      "model_slug": "ai-sage-gigachat3-5-432b-a28b-reasoning",
      "model_name": "GigaChat3.5-432B-A28B-Reasoning",
      "publisher": "ai-sage",
      "hf_repo": "ai-sage/GigaChat3.5-432B-A28B-Reasoning",
      "config_revision": "02f55379ac0b20abd9378791bffd6254725e6f5c",
      "parameters": 438085063424,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 269957533589,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.8194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.6145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.5462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.5462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 45202325504,
      "utilisation": 0.9417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.4777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.9832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 56597909504,
      "utilisation": 0.8843,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.8765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.8194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.821,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.3073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 56597909504,
      "utilisation": 0.8843,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 56597909504,
      "utilisation": 0.8843,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.3073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.9555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 65240981504,
      "utilisation": 0.8155,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 65240981504,
      "utilisation": 0.8155,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.5827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.6868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 7.3886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.6868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 7.3886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.9555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 4.9258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 3.6943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 2.4629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.8472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29554581504,
      "utilisation": 0.9236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 65240981504,
      "utilisation": 0.8155,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 65240981504,
      "utilisation": 0.8155,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.5968,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.5968,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29554581504,
      "utilisation": 1.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5554581504
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 45202325504,
      "utilisation": 0.9417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 45202325504,
      "utilisation": 0.9417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 45202325504,
      "utilisation": 0.9417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 45202325504,
      "utilisation": 0.9417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 65240981504,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 84144533504,
      "utilisation": 0.8765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "aleph-alpha-kolibri-1-bf16",
      "model_name": "Kolibri-1-BF16",
      "publisher": "Aleph-Alpha",
      "hf_repo": "Aleph-Alpha/Kolibri-1-BF16",
      "config_revision": "7a8f290e7858825c3cf5e4c447ba68345de9f1d3",
      "parameters": 78103074560,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 157319573504,
      "utilisation": 0.5462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0325-32b-instruct",
      "model_name": "OLMo-2-0325-32B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0325-32B-Instruct",
      "config_revision": "b96024342a77a69aa0dda815c3454a671f477463",
      "parameters": 32234279936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-2-0425-1b",
      "model_name": "OLMo-2-0425-1B",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMo-2-0425-1B",
      "config_revision": "a1847dff35000b4271fa70afc5db10fd29fedbdf",
      "parameters": 1484916736,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.3768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.9042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0353,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0353,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9475909104,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1005,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9475909104,
      "utilisation": 0.8614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.5719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9475909104,
      "utilisation": 0.8614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.5719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9475909104,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6287404208,
      "utilisation": 1.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 287404208
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7955617584,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.9369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11243268208,
      "utilisation": 0.7027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.5651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.7535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.3768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.3768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.3768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.3768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.2512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.1884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmo-3-7b-instruct",
      "model_name": "Olmo-3-7B-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/Olmo-3-7B-Instruct",
      "config_revision": "6e5971d9eba42665f5bd5a0fcf047f299ce1dccc",
      "parameters": 7298011136,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18084658288,
      "utilisation": 0.0628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "allenai-olmoe-1b-7b-0125-instruct",
      "model_name": "OLMoE-1B-7B-0125-Instruct",
      "publisher": "allenai",
      "hf_repo": "allenai/OLMoE-1B-7B-0125-Instruct",
      "config_revision": "b89a7c4bc24fb9e55ce2543c9458ce0ca5c4650e",
      "parameters": 6919161856,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.2881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.2161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.1921,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.1921,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18006511233,
      "utilisation": 0.9003,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.3457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.8642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.5762,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.2881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.8642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.8642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.2147,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.6914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.6914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.3073,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.2049,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.4321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.1042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 3.0366,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.1042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 3.0366,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.2147,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 2.0244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.5183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12146545568,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146545568
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14947091408,
      "utilisation": 0.9342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.9405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.6914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.6914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.3923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.3923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23584068915,
      "utilisation": 0.9827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30096262543,
      "utilisation": 0.627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.5762,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "altworld-hemmingway-1",
      "model_name": "Hemmingway-1",
      "publisher": "Altworld",
      "hf_repo": "Altworld/Hemmingway-1",
      "config_revision": "b987f1800da287b3016862948743a253fbf80553",
      "parameters": 26895998464,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55311261103,
      "utilisation": 0.1921,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.3447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 44713170944,
      "utilisation": 0.9315,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 57931820768,
      "utilisation": 0.9052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.8411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 57931820768,
      "utilisation": 0.9052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 57931820768,
      "utilisation": 0.9052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67831983968,
      "utilisation": 0.8479,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67831983968,
      "utilisation": 0.8479,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.8272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.6308,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.7274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 7.5003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.7274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 7.5003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67831983968,
      "utilisation": 0.8479,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67831983968,
      "utilisation": 0.8479,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.5727,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.5727,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 44713170944,
      "utilisation": 0.9315,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 44713170944,
      "utilisation": 0.9315,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 44713170944,
      "utilisation": 0.9315,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 44713170944,
      "utilisation": 0.9315,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67831983968,
      "utilisation": 0.9421,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80746966880,
      "utilisation": 0.8411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "anthracite-org-magnum-v4-72b",
      "model_name": "magnum-v4-72b",
      "publisher": "anthracite-org",
      "hf_repo": "anthracite-org/magnum-v4-72b",
      "config_revision": "3ea7aecd096b6323fcc4c01c6cef11364c75e6b5",
      "parameters": 72706203648,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.9975,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1247,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.5542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.885,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0739,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4848199075,
      "utilisation": 1.212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 848199075
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4848199075,
      "utilisation": 1.212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 848199075
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.885,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "apple-lensvlm-9b",
      "model_name": "LensVLM-9B",
      "publisher": "apple",
      "hf_repo": "apple/LensVLM-9B",
      "config_revision": "ac40d9667bd68df3ac9839d3f0dc7bdb93f5f92b",
      "parameters": 9409813744,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183416256512,
      "utilisation": 0.9553,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 243906529632,
      "utilisation": 0.9528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 285665380768,
      "utilisation": 0.9919,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 285665380768,
      "utilisation": 0.9919,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 422914353152,
      "utilisation": 0.979,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.9433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 7.064,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 141279586208,
      "utilisation": 0.883,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.2075,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.4717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183416256512,
      "utilisation": 0.9553,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 3.9244,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 422914353152,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.2075,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.2075,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 422914353152,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 14.128,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.766,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.766,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 141279586208,
      "utilisation": 0.7849,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 243906529632,
      "utilisation": 0.9034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.1037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 12.8436,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 35.3199,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 12.8436,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 35.3199,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 14.128,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 23.5466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 17.6599,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 11.7733,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 8.83,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 4.415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.766,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.766,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 5.8866,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.9433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.9433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.9433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 2.9433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.9622,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 141279586208,
      "utilisation": 1.4717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45279586208
    },
    {
      "model_slug": "arcee-ai-trinity-large-thinking",
      "model_name": "Trinity-Large-Thinking",
      "publisher": "arcee-ai",
      "hf_repo": "arcee-ai/Trinity-Large-Thinking",
      "config_revision": "dc6a99b6b74202880e31491a445a94884c97890c",
      "parameters": 398635286016,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 285665380768,
      "utilisation": 0.9919,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.2353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1046,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.9414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.9648,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.2824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.4707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.2353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.6836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.0883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.0883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.9961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.5648,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.5648,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1674,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.9056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 2.4904,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.9056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 2.4904,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.9961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.6602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9961429748,
      "utilisation": 1.2452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1961429748
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9961429748,
      "utilisation": 0.8301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14743493132,
      "utilisation": 0.9215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24609413918,
      "utilisation": 0.769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.5648,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.5648,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.3205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.3205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19295095630,
      "utilisation": 0.804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.9414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.9414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.9414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.9414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.6276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.4707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-21b-a3b-pt",
      "model_name": "ERNIE-4.5-21B-A3B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-21B-A3B-PT",
      "config_revision": "87db95487941cb39592ee0abca3b9155a6d19c5c",
      "parameters": 21948655808,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 45186278738,
      "utilisation": 0.1569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 187065447862,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 249376254349,
      "utilisation": 0.9741,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 249376254349,
      "utilisation": 0.8659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 249376254349,
      "utilisation": 0.8659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 322128534135,
      "utilisation": 0.7457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 2.5333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 6.08,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 152886524472,
      "utilisation": 0.9555,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.9,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.2667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 187065447862,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.3778,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 85599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 322128534135,
      "utilisation": 0.6292,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.9,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.9,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 322128534135,
      "utilisation": 0.6292,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 12.16,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.52,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.52,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 177450278205,
      "utilisation": 0.9858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 249376254349,
      "utilisation": 0.9236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.95,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 11.0545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 30.3999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 11.0545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 30.3999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 12.16,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 20.2666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 15.2,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 10.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 7.6,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 3.8,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.52,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.52,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.8624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121599663831,
      "utilisation": 0.8624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 5.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 2.5333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 2.5333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 2.5333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 2.5333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.6889,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121599663831,
      "utilisation": 1.2667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25599663831
    },
    {
      "model_slug": "baidu-ernie-4-5-300b-a47b-pt",
      "model_name": "ERNIE-4.5-300B-A47B-PT",
      "publisher": "baidu",
      "hf_repo": "baidu/ERNIE-4.5-300B-A47B-PT",
      "config_revision": "47bbaf1c9e356313c3cb2ded9078a494fb152f0e",
      "parameters": 300474051776,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 249376254349,
      "utilisation": 0.8659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.8931,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.4961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8079703914,
      "utilisation": 0.808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0662,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.917,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4553460044,
      "utilisation": 1.1384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 553460044
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.917,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4553460044,
      "utilisation": 1.1384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 553460044
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8079703914,
      "utilisation": 0.808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2481,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "bytedance-seed-ui-tars-1-5-7b",
      "model_name": "UI-TARS-1.5-7B",
      "publisher": "ByteDance-Seed",
      "hf_repo": "ByteDance-Seed/UI-TARS-1.5-7B",
      "config_revision": "683d002dd99d8f95104d31e70391a39348857f4e",
      "parameters": 8292166656,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cactus-compute-needle2",
      "model_name": "needle2",
      "publisher": "Cactus-Compute",
      "hf_repo": "Cactus-Compute/needle2",
      "config_revision": "98fbd955b0347e78059be0c253cc1ffa09b87bc7",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "cmsmanhattan-jirackultra-1b",
      "model_name": "JiRackUltra_1b",
      "publisher": "CMSManhattan",
      "hf_repo": "CMSManhattan/JiRackUltra_1b",
      "config_revision": "1a9946ec7311e84b7e80fb653b748a5e9b38d08d",
      "parameters": 1777088000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0967,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.9287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0967,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9879024768,
      "utilisation": 0.9879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.8935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.8935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9879024768,
      "utilisation": 0.9879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7574182016,
      "utilisation": 1.2624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1574182016
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7574182016,
      "utilisation": 0.9468,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10625070720,
      "utilisation": 0.8854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12256998016,
      "utilisation": 0.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.5804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "codellama-codellama-7b-hf",
      "model_name": "CodeLlama-7b-hf",
      "publisher": "codellama",
      "hf_repo": "codellama/CodeLlama-7b-hf",
      "config_revision": "6c284d1468fe6c413cf56183e69b194dcfa27fe6",
      "parameters": 6738546688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18574135936,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3182,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0663,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.6364,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3435276288,
      "utilisation": 0.8588,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3435276288,
      "utilisation": 0.8588,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.6364,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4036634624,
      "utilisation": 0.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.7956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.5304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.1326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0663,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-micro-vision-instruct",
      "model_name": "North-Micro-Vision-Instruct",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Micro-Vision-Instruct",
      "config_revision": "46b719694e3bad142f3e931774f4622f1024009e",
      "parameters": 2484847856,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6364473344,
      "utilisation": 0.0221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.3301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.22,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.22,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.1467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.7149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18954708992,
      "utilisation": 0.9477,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.3961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.3301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.9532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.1238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.1238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.7921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.7921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.3521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.2347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.1639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 3.2008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.1639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 3.2008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 2.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.6004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12803086336,
      "utilisation": 1.0669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 803086336
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12803086336,
      "utilisation": 0.8002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26807023616,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.7921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.7921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.4494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23407310848,
      "utilisation": 0.9753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.7149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.7149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.7149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34313463808,
      "utilisation": 0.7149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.8801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "coherelabs-north-mini-code-1-0",
      "model_name": "North-Mini-Code-1.0",
      "publisher": "CohereLabs",
      "hf_repo": "CohereLabs/North-Mini-Code-1.0",
      "config_revision": "d11e61a842617a22dc328552fa5bb86231ee4f37",
      "parameters": 30484303872,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63370651648,
      "utilisation": 0.22,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0968,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.9289,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0968,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9880516608,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.9661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.8938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.9661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.8938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9880516608,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7575320576,
      "utilisation": 1.2626,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1575320576
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7575320576,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10626693120,
      "utilisation": 0.8856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12259096576,
      "utilisation": 0.7662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.5806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.7741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-6-7b-instruct",
      "model_name": "deepseek-coder-6.7b-instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-6.7b-instruct",
      "config_revision": "e5d64addd26a6a1db0f9b863abf6ee3141936807",
      "parameters": 6740512768,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18578077696,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-7b-instruct-v1-5",
      "model_name": "deepseek-coder-7b-instruct-v1.5",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
      "config_revision": "2a050a4c59d687a85324d32e147517992117ed30",
      "parameters": 6910365696,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1269,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0752,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.8878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.4511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-coder-v2-lite-instruct",
      "model_name": "DeepSeek-Coder-V2-Lite-Instruct",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
      "config_revision": "e434a23f91ba5b4923cf6c9d9a238eb4a08e3a11",
      "parameters": 15706484224,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.4189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.0642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272433647044,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272433647044,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421566872133,
      "utilisation": 0.9758,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 5.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 13.6217,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.7027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 4.2568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.4189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 7.5676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490358101606,
      "utilisation": 0.9577,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 4.2568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 4.2568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490358101606,
      "utilisation": 0.9577,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 27.2434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 3.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 3.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.5135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.1284,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 24.7667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 68.1084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 24.7667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 68.1084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 27.2434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 45.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 34.0542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 22.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 17.0271,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 8.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 3.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 3.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.9322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 1.9322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 11.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 5.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 5.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 5.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 5.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 3.7838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272433647044,
      "utilisation": 2.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176433647044
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1",
      "model_name": "DeepSeek-R1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1",
      "config_revision": "56d4cbbb4d29f4355bab4b9a39ccb717a14ad5ad",
      "parameters": 684489845504,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272433647044,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421592372805,
      "utilisation": 0.9759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 13.6225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 7.5681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528",
      "model_name": "DeepSeek-R1-0528",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528",
      "config_revision": "4236a6af538feda4548eca9ab308586007567f52",
      "parameters": 684531386000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.9198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8734241792,
      "utilisation": 0.8734,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.9744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5208679424,
      "utilisation": 1.3022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1208679424
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.9744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5208679424,
      "utilisation": 1.3022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1208679424
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8734241792,
      "utilisation": 0.8734,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859454976,
      "utilisation": 0.9824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.8932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717860864,
      "utilisation": 0.6699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-0528-qwen3-8b",
      "model_name": "DeepSeek-R1-0528-Qwen3-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
      "config_revision": "6e8885a6ff5c1dc5201574c8fd700323f23c25fa",
      "parameters": 8190735360,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-70b",
      "model_name": "DeepSeek-R1-Distill-Llama-70B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
      "config_revision": "b1c0b44b4369b597ad119a196caf79a9c40e141e",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-llama-8b",
      "model_name": "DeepSeek-R1-Distill-Llama-8B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
      "config_revision": "6a6f4aa4197940add57724a7707d069478df56b1",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-1-5b",
      "model_name": "DeepSeek-R1-Distill-Qwen-1.5B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
      "config_revision": "ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562",
      "parameters": 1777088000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1248,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.8877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1775,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-14b",
      "model_name": "DeepSeek-R1-Distill-Qwen-14B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
      "config_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
      "parameters": 14770033664,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.7868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19173353472,
      "utilisation": 0.9587,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.5901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.7134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.8287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.5901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.5901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.5044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.3677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 3.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.3677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 3.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.5044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 2.5074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.8805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15044155392,
      "utilisation": 1.2537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3044155392
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15044155392,
      "utilisation": 0.9403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29832169472,
      "utilisation": 0.9323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22797350912,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.7868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.7868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.7868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37766899712,
      "utilisation": 0.7868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.7134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-32b",
      "model_name": "DeepSeek-R1-Distill-Qwen-32B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
      "config_revision": "711ad2ea6aa40cfca18895e8aca02ab92df1a746",
      "parameters": 32763876352,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.8254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.4585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.9368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.8517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4242225152,
      "utilisation": 1.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 242225152
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.8517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4242225152,
      "utilisation": 1.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 242225152
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.9368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-r1-distill-qwen-7b",
      "model_name": "DeepSeek-R1-Distill-Qwen-7B",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
      "config_revision": "916b56a44061fd5cd7d6a8fb632557ed4f724f60",
      "parameters": 7615616512,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1269,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0752,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.8878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.4511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite",
      "model_name": "DeepSeek-V2-Lite",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite",
      "config_revision": "604d5664dddd88a0433dbae533b7fe9472482de0",
      "parameters": 15706484224,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1269,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0752,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.8878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.5075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.0634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.8186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 8910009391,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7274571721,
      "utilisation": 1.2124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1274571721
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7274571721,
      "utilisation": 0.9093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10696621971,
      "utilisation": 0.8914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13953754137,
      "utilisation": 0.8721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.5549,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.2304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17756686630,
      "utilisation": 0.7399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.6767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.4511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.3383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v2-lite-chat",
      "model_name": "DeepSeek-V2-Lite-Chat",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V2-Lite-Chat",
      "config_revision": "85864749cd611b4353ce1decdb286193298f64c7",
      "parameters": 15706484224,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 32481515590,
      "utilisation": 0.1128,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421592372805,
      "utilisation": 0.9759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 13.6225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 7.5681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3",
      "model_name": "DeepSeek-V3",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3",
      "config_revision": "e815299b0bcbac849fa540c768ef21845365c9eb",
      "parameters": 684531386000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421592372805,
      "utilisation": 0.9759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 13.6225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 7.5681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-0324",
      "model_name": "DeepSeek-V3-0324",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3-0324",
      "config_revision": "e9b33add76883f293d6bf61f6bd89b497e80e335",
      "parameters": 684531386000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421592372805,
      "utilisation": 0.9759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 13.6225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 7.5681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1",
      "model_name": "DeepSeek-V3.1",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1",
      "config_revision": "c0781d039fb7a1ba2abc4add0bdc293e92d2b8db",
      "parameters": 684531386000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 421592372805,
      "utilisation": 0.9759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 13.6225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.7028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 7.5681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 4.257,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 490387777098,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.5136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.0091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 24.7682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 68.1125,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 27.245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 45.4083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 34.0563,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 22.7042,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 17.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 8.5141,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.4056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 1.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 11.3521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 5.676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 3.784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272450097080,
      "utilisation": 2.838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176450097080
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-1-terminus",
      "model_name": "DeepSeek-V3.1-Terminus",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.1-Terminus",
      "config_revision": "19510d6dc61f79dbd925bd51ee8a9081c509a4b6",
      "parameters": 684531386000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272450097080,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.4214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.066,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272904305094,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272904305094,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 422226097572,
      "utilisation": 0.9774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 5.6855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 13.6452,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.7057,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 4.2641,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.8428,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.4214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 7.5807,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 491104308216,
      "utilisation": 0.9592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 4.2641,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 4.2641,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 491104308216,
      "utilisation": 0.9592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 27.2904,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 3.4113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 3.4113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.5161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.0108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 24.8095,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 68.2261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 24.8095,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 68.2261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 27.2904,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 45.4841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 34.113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 22.742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 17.0565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 8.5283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 3.4113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 3.4113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.9355,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 1.9355,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 11.371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 5.6855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 5.6855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 5.6855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 5.6855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 3.7903,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272904305094,
      "utilisation": 2.8428,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176904305094
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2",
      "model_name": "DeepSeek-V3.2",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2",
      "config_revision": "a7e62ac04ecb2c0a54d736dc46601c5606cf10a6",
      "parameters": 685355329792,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272904305094,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.4215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.0661,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272920775361,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272920775361,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 422251629606,
      "utilisation": 0.9774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 5.6858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 13.646,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 252920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.7058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 4.2644,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.8429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.4215,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 7.5811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 491134020204,
      "utilisation": 0.9592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 4.2644,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 4.2644,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 491134020204,
      "utilisation": 0.9592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 27.2921,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 3.4115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 3.4115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.5162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.0108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.1322,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 24.811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 68.2302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 24.811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 68.2302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 27.2921,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 45.4868,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 34.1151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 22.7434,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 17.0575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 8.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 240920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 3.4115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 3.4115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.9356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 1.9356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 131920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 11.3717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 5.6858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 5.6858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 5.6858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 5.6858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 224920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 3.7906,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 272920775361,
      "utilisation": 2.8429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176920775361
    },
    {
      "model_slug": "deepseek-ai-deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V3.2-Exp",
      "config_revision": "194c67e12b1b0d6df0ef373ddcf215bc84027409",
      "parameters": 685396921376,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 272920775361,
      "utilisation": 0.9476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.0455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 267654513088,
      "utilisation": 0.9294,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 267654513088,
      "utilisation": 0.9294,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 426074725664,
      "utilisation": 0.9863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 5.5761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 13.3827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 247654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.6728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 4.1821,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 203654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.7881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 7.4348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 449445272864,
      "utilisation": 0.8778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 4.1821,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 203654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 4.1821,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 203654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 449445272864,
      "utilisation": 0.8778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 26.7655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 257654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 3.3457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 3.3457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 87654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 267654513088,
      "utilisation": 0.9913,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 24.3322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 66.9136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 263654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 24.3322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 66.9136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 263654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 26.7655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 257654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 44.6091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 261654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 33.4568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 22.3045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 255654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 16.7284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 251654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 8.3642,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 3.3457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 3.3457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.8983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 1.8983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 11.1523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 243654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 5.5761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 5.5761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 5.5761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 5.5761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 3.7174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 267654513088,
      "utilisation": 2.7881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171654513088
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4.1-Flash",
      "config_revision": "dba1be0a40aa45a94ad051997016db3960a90277",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 267654513088,
      "utilisation": 0.9294,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 175879884800,
      "utilisation": 0.916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 238821404672,
      "utilisation": 0.9329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 238821404672,
      "utilisation": 0.8292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 238821404672,
      "utilisation": 0.8292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 309009166336,
      "utilisation": 0.7153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.2048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 5.2915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 85830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 143740956672,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.6536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.1024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 175879884800,
      "utilisation": 0.916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.9397,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 309009166336,
      "utilisation": 0.6035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.6536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.6536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 309009166336,
      "utilisation": 0.6035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 10.583,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.3229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.3229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 175879884800,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 238821404672,
      "utilisation": 0.8845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 9.6209,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 26.4575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 9.6209,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 26.4575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 10.583,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 17.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 13.2288,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 8.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 6.6144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 3.3072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.3229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.3229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.7506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 105830010880,
      "utilisation": 0.7506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 4.4096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.2048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.2048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.2048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 2.2048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.4699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 105830010880,
      "utilisation": 1.1024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9830010880
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash",
      "config_revision": "60d8d70770c6776ff598c94bb586a859a38244f1",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 238821404672,
      "utilisation": 0.8292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 187519449728,
      "utilisation": 0.9767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 250561449728,
      "utilisation": 0.9788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 250561449728,
      "utilisation": 0.87,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 250561449728,
      "utilisation": 0.87,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 324167449728,
      "utilisation": 0.7504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 2.5268,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 6.0643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 152939449728,
      "utilisation": 0.9559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.8951,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.2634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 187519449728,
      "utilisation": 0.9767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.369,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 85285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 324167449728,
      "utilisation": 0.6331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.8951,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.8951,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 324167449728,
      "utilisation": 0.6331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 12.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.5161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.5161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 177791449728,
      "utilisation": 0.9877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 250561449728,
      "utilisation": 0.928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.9475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 11.0259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 30.3214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 11.0259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 30.3214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 12.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 20.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 15.1607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 10.1071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 7.5803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 3.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.5161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.5161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.8602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121285449728,
      "utilisation": 0.8602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 5.0536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 2.5268,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 2.5268,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 2.5268,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 2.5268,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.6845,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121285449728,
      "utilisation": 1.2634,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25285449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-0731",
      "model_name": "DeepSeek-V4-Flash-0731",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-0731",
      "config_revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 250561449728,
      "utilisation": 0.87,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 188133324728,
      "utilisation": 0.9799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 251382699728,
      "utilisation": 0.982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 251382699728,
      "utilisation": 0.8729,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 251382699728,
      "utilisation": 0.8729,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 325230824728,
      "utilisation": 0.7528,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 2.535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 6.0841,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 153439574728,
      "utilisation": 0.959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.9013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.2675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 188133324728,
      "utilisation": 0.9799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.38,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 85681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 325230824728,
      "utilisation": 0.6352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.9013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.9013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 57681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 325230824728,
      "utilisation": 0.6352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 12.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178373324728,
      "utilisation": 0.991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 251382699728,
      "utilisation": 0.931,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 11.0619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 30.4204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 11.0619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 30.4204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 12.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 20.2802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 15.2102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 10.1401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 7.6051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 3.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.863,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 121681449728,
      "utilisation": 0.863,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 5.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 2.535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 2.535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 2.535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 2.535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.69,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 121681449728,
      "utilisation": 1.2675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25681449728
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-flash-vision-exp",
      "model_name": "DeepSeek-V4-Flash-Vision-Exp",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
      "config_revision": "6821d6ad3681a4b137b066b76094fa82ebd0a380",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 251382699728,
      "utilisation": 0.8729,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.2726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.3467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 29.0894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 561788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.6362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 421788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 6.0603,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 485788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 16.1608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 545788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 58.1789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 571788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 401788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.1548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 311788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 52.8899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 570788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 145.4472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 577788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 52.8899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 570788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 145.4472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 577788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 58.1789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 571788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 440788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 440788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 8.0804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 509788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 6.0603,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 485788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro",
      "model_name": "DeepSeek-V4-Pro",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro",
      "config_revision": "b5968e9190ef611bbf34a7229255be88a0e937c1",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.2726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.3467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 29.0894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 561788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.6362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 421788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 6.0603,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 485788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 16.1608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 545788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 9.0904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 517788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 69788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 58.1789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 571788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 3.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 401788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.1548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 311788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.5452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 453788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 52.8899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 570788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 145.4472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 577788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 52.8899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 570788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 145.4472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 577788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 58.1789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 571788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 96.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 575788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 72.7236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 573788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 48.4824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 569788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 36.3618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 565788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 18.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 549788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 7.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 501788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 440788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 4.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 440788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 24.2412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 557788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 12.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 533788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 8.0804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 509788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 6.0603,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 485788719104
    },
    {
      "model_slug": "deepseek-ai-deepseek-v4-pro-0813",
      "model_name": "DeepSeek-V4-Pro-0813",
      "publisher": "deepseek-ai",
      "hf_repo": "deepseek-ai/DeepSeek-V4-Pro-0813",
      "config_revision": "72e1d3230f6c080a530b0a1d46f8eb4602340597",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 581788719104,
      "utilisation": 2.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 293788719104
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.3729,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.2797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.2486,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.2486,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.1657,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.8199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19581208576,
      "utilisation": 0.9791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.4475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.615,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.7458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.3729,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.8619,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.1398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.615,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.615,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.1398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.5268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.8949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.8949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.3978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.2652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 3.8171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 3.8171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.5268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 2.5447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.9085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15268222976,
      "utilisation": 1.2724,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3268222976
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15268222976,
      "utilisation": 0.9543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 31028504576,
      "utilisation": 0.9696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.8949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.8949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5078,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.5078,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23473289216,
      "utilisation": 0.9781,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.8199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.8199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.8199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39356860416,
      "utilisation": 0.8199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.9944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.7458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-2-9-1-yi-1-5-34b",
      "model_name": "dolphin-2.9.1-yi-1.5-34b",
      "publisher": "dphn",
      "hf_repo": "dphn/dolphin-2.9.1-yi-1.5-34b",
      "config_revision": "0141cba238d0faad09bc240ea7af14c9ea5aec44",
      "parameters": 34388917248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71595657216,
      "utilisation": 0.2486,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1926,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18903783424,
      "utilisation": 0.9452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.7554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6846,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "dphn-dolphin-mistral-24b-venice-edition",
      "model_name": "Dolphin-Mistral-24B-Venice-Edition",
      "publisher": "dphn",
      "hf_repo": "dphn/Dolphin-Mistral-24B-Venice-Edition",
      "config_revision": "b23ff40b4a25e4f93de432bd896d4464479a9f79",
      "parameters": 24011361280,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.3701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.2776,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.2468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.2468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.1645,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.7969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18538993880,
      "utilisation": 0.9269,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.4442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.5977,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.7403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.3701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.8272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.1388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.5977,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.5977,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.1388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.8883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.8883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.3948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.2632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.5552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 3.7237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 3.7237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 2.4824,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14894618880,
      "utilisation": 1.2412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2894618880
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14894618880,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29778368880,
      "utilisation": 0.9306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.8883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.8883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22520243880,
      "utilisation": 0.9383,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.7969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.7969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.7969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38252743880,
      "utilisation": 0.7969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.7403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "edge0-edge0-35b-a3b-preview",
      "model_name": "Edge0-35B-A3B-preview",
      "publisher": "Edge0",
      "hf_repo": "Edge0/Edge0-35B-A3B-preview",
      "config_revision": "1ff9f4478890faec0368c5463b621d1036d5b518",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71065243880,
      "utilisation": 0.2468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-gpt-neox-20b",
      "model_name": "gpt-neox-20b",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/gpt-neox-20b",
      "config_revision": "c292233c833e336628618a88a648727eb3dff0a7",
      "parameters": 20739117584,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-14m",
      "model_name": "pythia-14m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-14m",
      "config_revision": "cf967c0a9a04383db6f7b1108d86b2962634b4ac",
      "parameters": 14067712,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m",
      "model_name": "pythia-160m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m",
      "config_revision": "50f5173d932e8e61f858120bcb800b97af589f46",
      "parameters": 212654688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-160m-deduped",
      "model_name": "pythia-160m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-160m-deduped",
      "config_revision": "582159a2dfe3e712a8d47ae83dec95ae3bde8e7e",
      "parameters": 212654688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-410m",
      "model_name": "pythia-410m",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-410m",
      "config_revision": "9879c9b5f8bea9051dcb0e68dff21493d67e9d4f",
      "parameters": 505997504,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "eleutherai-pythia-70m-deduped",
      "model_name": "pythia-70m-deduped",
      "publisher": "EleutherAI",
      "hf_repo": "EleutherAI/pythia-70m-deduped",
      "config_revision": "e93a9faa9c77e5d09219f6c868bfc7a1bd65593c",
      "parameters": 95592496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18592598246,
      "utilisation": 0.9296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.4455,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.8296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.99,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-35b-a3b-distill",
      "model_name": "Qwen3.8-35B-A3B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-35B-A3B-Distill",
      "config_revision": "bcc2dbe2f21b213625df2dc1a5a690212373af07",
      "parameters": 35107181936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0382,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0227,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0612,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1529,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2719,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0191,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1529,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1529,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0191,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.9788,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1223,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1223,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0544,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8898,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3905359136,
      "utilisation": 0.9763,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8898,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3905359136,
      "utilisation": 0.9763,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.9788,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.9554,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5732492576,
      "utilisation": 0.7166,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.8156,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.6117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.3059,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1223,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1223,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0694,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.0694,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.4078,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.2039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.1359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-4b-distill",
      "model_name": "Qwen3.8-4B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-4B-Distill",
      "config_revision": "c83cb7aa2999d2f35c43e9ae0634a30eb8985a1e",
      "parameters": 4659865088,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9787532576,
      "utilisation": 0.034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1017,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0763,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0452,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.9765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1017,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.5425,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0381,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0381,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8680813888,
      "utilisation": 0.8681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1085,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0723,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.9916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.9916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8680813888,
      "utilisation": 0.8681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764456768,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907972928,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.1385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.2034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "empero-ai-qwen3-8-9b-distill",
      "model_name": "Qwen3.8-9B-Distill",
      "publisher": "empero-ai",
      "hf_repo": "empero-ai/Qwen3.8-9B-Distill",
      "config_revision": "0934f3d2327ff2df2197495278c4c46ae5a56bd9",
      "parameters": 9653104368,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529233728,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2582,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18716392448,
      "utilisation": 0.9358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4131,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.8022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.1291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.1291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.4881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.3528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 3.7201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.3528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 3.7201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.4881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.9181,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-2-0-llm-31b-it",
      "model_name": "OTel-2.0-LLM-31B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-2.0-LLM-31B-IT",
      "config_revision": "522937ff14a94c47b2d2960b95e474ec7ab40aa2",
      "parameters": 31273088876,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-27b-it",
      "model_name": "OTel-LLM-27B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-27B-IT",
      "config_revision": "62a522b2dcb833a5d2ef1141b079b35e4b8f1898",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "farbodtavakkoli-otel-llm-e4b-it",
      "model_name": "OTel-LLM-E4B-IT",
      "publisher": "farbodtavakkoli",
      "hf_repo": "farbodtavakkoli/OTel-LLM-E4B-IT",
      "config_revision": "12ffb1ef5812f53ea2c7732b4cc3703c2d2171a9",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18098272256,
      "utilisation": 0.9049,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.8086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.9632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "flywheel-ai-automotive",
      "model_name": "automotive",
      "publisher": "flywheel-ai",
      "hf_repo": "flywheel-ai/automotive",
      "config_revision": "f75f112dabbda2c434d00af21e99337159604ed3",
      "parameters": 34660610688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1064,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.0945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.0945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.7627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.4254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.7563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.0532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.4254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.4254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.0532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9541846016,
      "utilisation": 0.9542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.3403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.3403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.9829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.6742,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.9829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.6742,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9541846016,
      "utilisation": 0.9542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6696682496,
      "utilisation": 1.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696682496
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6696682496,
      "utilisation": 0.8371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10811933696,
      "utilisation": 0.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.9534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.3403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.3403,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1931,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.1931,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15254537216,
      "utilisation": 0.6356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.3782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.2836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-12b-it",
      "model_name": "gemma-4-12B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-12B-it",
      "config_revision": "707f0a3b8a3c7ad586ed01e27eafbad8a27dd0f7",
      "parameters": 11959730224,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 27227964416,
      "utilisation": 0.0945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.2035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.5904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 19410363392,
      "utilisation": 0.9705,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.3257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.5428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.7871,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.0873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.6513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.6513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.2895,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.193,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 2.7182,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 2.7182,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.0873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.8121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10872792064,
      "utilisation": 1.3591,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2872792064
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10872792064,
      "utilisation": 0.9061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15851029504,
      "utilisation": 0.9907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.8855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.6513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.6513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.3695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.3695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22196896768,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.5904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.5904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.5904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 28336948224,
      "utilisation": 0.5904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.7237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.5428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-26b-a4b-it",
      "model_name": "gemma-4-26B-A4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-26B-A4B-it",
      "config_revision": "4d7ae4984b7db7de8f8457170b3f1a419ee76d52",
      "parameters": 25805936206,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 52104889344,
      "utilisation": 0.1809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2582,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18716392448,
      "utilisation": 0.9358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4131,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.8022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.1291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.5706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.1291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.4881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.3672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.5164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.3528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 3.7201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.3528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 3.7201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.4881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 2.4801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.8601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14880594944,
      "utilisation": 1.24,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2880594944
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14880594944,
      "utilisation": 0.93,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28879182848,
      "utilisation": 0.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.8263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.4688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22171741184,
      "utilisation": 0.9238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36521360384,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.9181,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-31b-it",
      "model_name": "gemma-4-31B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-31B-it",
      "config_revision": "842da3794eaa0b77d5f08bae87a17459d91ff475",
      "parameters": 31273088876,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66103983104,
      "utilisation": 0.2295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0435,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.2318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.5562,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1159,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0217,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0217,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.6321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0412,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.5747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 3854449548,
      "utilisation": 0.9636,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.5747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 3854449548,
      "utilisation": 0.9636,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.6321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5080810294,
      "utilisation": 0.8468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6321259780,
      "utilisation": 0.7902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.6953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.3476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.4635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.2318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.2318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.2318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.2318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.1159,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e2b-it",
      "model_name": "gemma-4-E2B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E2B-it",
      "config_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
      "parameters": 5123178051,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11124239203,
      "utilisation": 0.0386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0662,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.3532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.8477,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2649,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1766,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.4709,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2649,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2649,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.9457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0942,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0628,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.8597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4120618642,
      "utilisation": 1.0302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120618642
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.8597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4120618642,
      "utilisation": 1.0302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120618642
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.9457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5862781237,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7520984189,
      "utilisation": 0.9401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9457053580,
      "utilisation": 0.5911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.5298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.7064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.3532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.3532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.3532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.3532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.2355,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.1766,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "google-gemma-4-e4b-it",
      "model_name": "gemma-4-E4B-it",
      "publisher": "google",
      "hf_repo": "google/gemma-4-E4B-it",
      "config_revision": "ee0ef6023621cff504d758262d4e04895a5af4a2",
      "parameters": 7996156490,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16953450289,
      "utilisation": 0.0589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "gryphe-mythomax-l2-13b",
      "model_name": "MythoMax-L2-13b",
      "publisher": "Gryphe",
      "hf_repo": "Gryphe/MythoMax-L2-13b",
      "config_revision": "58e77dd48a65176f97f6f376c93efe9caad9c130",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.2929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.2197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.6372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18289341921,
      "utilisation": 0.9145,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.3515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.8786,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.5858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.2929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.8496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1098,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.8786,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.8786,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1098,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.2329,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.7029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.7029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.3124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.2083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.4393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.1208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 3.0822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.1208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 3.0822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.2329,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 2.0548,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.5411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12328994686,
      "utilisation": 1.0274,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 328994686
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15177514047,
      "utilisation": 0.9486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.9558,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.7029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.7029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.3988,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.3988,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23962443506,
      "utilisation": 0.9984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.6372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.6372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.6372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30586191408,
      "utilisation": 0.6372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.781,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.5858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-27b",
      "model_name": "Holo4-27B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-27B",
      "config_revision": "4de03cfeee1d930b616bfd4817570293f29ad8b9",
      "parameters": 27356728560,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 56233124433,
      "utilisation": 0.1953,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18592598246,
      "utilisation": 0.9296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.4455,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.8296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.99,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "hcompany-holo4-35b-a3b",
      "model_name": "Holo4-35B-A3B",
      "publisher": "Hcompany",
      "hf_repo": "Hcompany/Holo4-35B-A3B",
      "config_revision": "508458727831368da85d53a8055c2b0f923ed4f3",
      "parameters": 35107181936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3019,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.6037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5488,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3899598848,
      "utilisation": 0.9749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5488,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3899598848,
      "utilisation": 0.9749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.6037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b",
      "model_name": "SmolLM2-1.7B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B",
      "config_revision": "effd688a12921b4cc83e3312b6feb579f70f9c71",
      "parameters": 1711376384,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3019,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.6037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5488,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3899598848,
      "utilisation": 0.9749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5488,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3899598848,
      "utilisation": 0.9749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.6037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4338427904,
      "utilisation": 0.7231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.5031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.2515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.1258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.0629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-1-7b-instruct",
      "model_name": "SmolLM2-1.7B-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-1.7B-Instruct",
      "config_revision": "31b70e2e869a7173562077fd711b654946d38674",
      "parameters": 1711376384,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6037121024,
      "utilisation": 0.021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0051,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0137,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0026,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0026,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0049,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.3292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.3292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0137,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m",
      "model_name": "SmolLM2-135M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M",
      "config_revision": "93efa2f097d58c2a74874c7e644dbc9b0cee75a2",
      "parameters": 134515008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0051,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0137,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0026,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0026,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0049,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.3292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.3292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.2194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.1097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0137,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-135m-instruct",
      "model_name": "SmolLM2-135M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-135M-Instruct",
      "config_revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac",
      "parameters": 134515008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1316695296,
      "utilisation": 0.0046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0076,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m",
      "model_name": "SmolLM2-360M",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M",
      "config_revision": "f8027fd0eaeea54caa13c31d31b9fdc459c38b49",
      "parameters": 361821120,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0076,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1778,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1956,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm2-360m-instruct",
      "model_name": "SmolLM2-360M-Instruct",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM2-360M-Instruct",
      "config_revision": "a10cc1512eabd3dde888204e902eca88bddb4951",
      "parameters": 361821120,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1955911424,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0187,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.4043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2246,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0299,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b",
      "model_name": "SmolLM3-3B",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B",
      "config_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
      "parameters": 3075098624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0187,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.4043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2246,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0299,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggingfacetb-smollm3-3b-base",
      "model_name": "SmolLM3-3B-Base",
      "publisher": "HuggingFaceTB",
      "hf_repo": "HuggingFaceTB/SmolLM3-3B-Base",
      "config_revision": "d78a42f79198603e614095753484a04c10c2b940",
      "parameters": 3075098624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "huggyllama-llama-7b",
      "model_name": "llama-7b",
      "publisher": "huggyllama",
      "hf_repo": "huggyllama/llama-7b",
      "config_revision": "4782ad278652c7c71b72204d462d6d01eaaf7549",
      "parameters": 6738417664,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0382,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0286,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0255,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0255,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.017,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0458,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1146,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0764,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0382,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2037,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0143,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1146,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1146,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0143,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.7334,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0917,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0917,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0272,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3570140832,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3570140832,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.7334,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4342789792,
      "utilisation": 0.7238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.9167,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.6111,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.4584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.2292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0917,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0917,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.3056,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.1019,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0764,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-0-h-micro",
      "model_name": "granite-4.0-h-micro",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.0-h-micro",
      "config_revision": "d5f01a3ea75f088947be3aae039f4ad52837dfde",
      "parameters": 3191396096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7333689280,
      "utilisation": 0.0255,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.3161,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.237,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.2107,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.2107,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.1405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.7005,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19295825120,
      "utilisation": 0.9648,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.3793,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.9482,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.6321,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.3161,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.1185,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.9482,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.9482,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.1185,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.3671,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.7585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.7585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.3371,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.2248,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4741,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.2428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 3.4178,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.2428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 3.4178,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.3671,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 2.2785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.7089,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671263456,
      "utilisation": 1.1393,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671263456
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671263456,
      "utilisation": 0.8545,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26631662816,
      "utilisation": 0.8322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.7585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.7585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.4304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23440846048,
      "utilisation": 0.9767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.7005,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.7005,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.7005,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33622453472,
      "utilisation": 0.7005,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.6321,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-30b",
      "model_name": "granite-4.1-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-30b",
      "config_revision": "4fae6278f7132abf5e971f9de49ebbad09c54cce",
      "parameters": 28865728512,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60683579168,
      "utilisation": 0.2107,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0344,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.4398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.055,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3715805184,
      "utilisation": 0.929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3715805184,
      "utilisation": 0.929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.8941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5364445184,
      "utilisation": 0.6706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.7329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.5497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.2749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.3665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.1222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-3b",
      "model_name": "granite-4.1-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-3b",
      "config_revision": "c0650403e44e78ec0262dab1c90914c65b196c4e",
      "parameters": 3402836480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8795254784,
      "utilisation": 0.0305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1028,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0771,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0457,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.9865,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1233,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2055,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1028,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358654112,
      "utilisation": 0.9359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0731,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358654112,
      "utilisation": 0.8508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5554485920,
      "utilisation": 1.3886,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1554485920
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358654112,
      "utilisation": 0.8508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5554485920,
      "utilisation": 1.3886,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1554485920
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358654112,
      "utilisation": 0.9359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554485920,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490091680,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487787680,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1399,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.1399,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.274,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.2055,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-1-8b",
      "model_name": "granite-4.1-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.1-8b",
      "config_revision": "1504002f650e656a0a3789d99574df12e3e94ed0",
      "parameters": 8791592960,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729595040,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.3203,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.2403,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.2136,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.2136,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.1424,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.7096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19527039712,
      "utilisation": 0.9764,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.3844,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.961,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.6407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.3203,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.9461,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.1201,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.961,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.961,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.1201,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.3806,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.7688,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.7688,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.3417,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.2278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.2551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 3.4515,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.2551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 3.4515,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.3806,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 2.301,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.7258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13806140128,
      "utilisation": 1.1505,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1806140128
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13806140128,
      "utilisation": 0.8629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26968849120,
      "utilisation": 0.8428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.7688,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.7688,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4362,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.4362,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23723440864,
      "utilisation": 0.9885,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.7096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.7096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.7096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34059188960,
      "utilisation": 0.7096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.8542,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.6407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-30b",
      "model_name": "granite-4.2-30b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-30b",
      "config_revision": "70fc514076785017d93087f3d8b0676426b6b355",
      "parameters": 29276770304,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61505665984,
      "utilisation": 0.2136,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0458,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0344,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0204,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.4397,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.055,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1374,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0458,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2443,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0172,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1374,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1374,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0172,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0489,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0326,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0687,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7995,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3715100800,
      "utilisation": 0.9288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7995,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3715100800,
      "utilisation": 0.9288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.894,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5363740800,
      "utilisation": 0.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.7329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.5497,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.2748,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0624,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0624,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.3664,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.1221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-3b",
      "model_name": "granite-4.2-3b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-3b",
      "config_revision": "b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2",
      "parameters": 3659737600,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8794550400,
      "utilisation": 0.0305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1028,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0771,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0457,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.9865,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1233,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2055,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1028,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.3083,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358657664,
      "utilisation": 0.9359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0731,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1541,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358657664,
      "utilisation": 0.8508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5554489472,
      "utilisation": 1.3886,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1554489472
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358657664,
      "utilisation": 0.8508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5554489472,
      "utilisation": 1.3886,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1554489472
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9358657664,
      "utilisation": 0.9359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5554489472,
      "utilisation": 0.9257,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7490095232,
      "utilisation": 0.9363,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.9573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11487791232,
      "utilisation": 0.718,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.6165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2466,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1399,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.1399,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.8221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.411,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.274,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.2055,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-4-2-8b",
      "model_name": "granite-4.2-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-4.2-8b",
      "config_revision": "41a8a2d41c54ef4a71741b3e62604f4caaec9295",
      "parameters": 8791592960,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19729598592,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0722,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0642,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0642,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.9243,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1155,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2888,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5135,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0361,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2888,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2888,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0361,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8847643744,
      "utilisation": 0.8848,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1027,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0685,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1444,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9842,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5221894584,
      "utilisation": 1.3055,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1221894584
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9842,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5221894584,
      "utilisation": 1.3055,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1221894584
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8847643744,
      "utilisation": 0.8848,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5221894584,
      "utilisation": 0.8703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7939642464,
      "utilisation": 0.9925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.9022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10826444672,
      "utilisation": 0.6767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.5777,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1311,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.2568,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.1926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-granite-granite-guardian-3-3-8b",
      "model_name": "granite-guardian-3.3-8b",
      "publisher": "ibm-granite",
      "hf_repo": "ibm-granite/granite-guardian-3.3-8b",
      "config_revision": "b3421eda4ba6fc9f9a71121d7e62de08827469a4",
      "parameters": 8170864640,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18486319232,
      "utilisation": 0.0642,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ibm-research-powermoe-3b",
      "model_name": "PowerMoE-3b",
      "publisher": "ibm-research",
      "hf_repo": "ibm-research/PowerMoE-3b",
      "config_revision": "13fcb5a98001438bed01cf1ac4b423751dc4c2ea",
      "parameters": 3374286336,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1715,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0214,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0357,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.3429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.3118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.8573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.3118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.8573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.3429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.5716,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.4287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2858,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.1429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0357,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-0-9b",
      "model_name": "K2-Horizon-0.9B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-0.9B",
      "config_revision": "9fa6faa55fe1c9eb008bb241cc6fb7e4536d0e91",
      "parameters": 1078285824,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3429341184,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.2528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.6068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3371,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.7393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4099890176,
      "utilisation": 1.025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99890176
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4099890176,
      "utilisation": 1.025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99890176
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.7393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 5650564096,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.9241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 7393191936,
      "utilisation": 0.6161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.7584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.3792,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.5056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.2528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.2528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.2528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.2528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1685,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.1264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-3-7b",
      "model_name": "K2-Horizon-3.7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-3.7B",
      "config_revision": "242d92f6f2c58342a8b184e6c4e705491436f81a",
      "parameters": 5058255360,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12135131136,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 189376585728,
      "utilisation": 0.9863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 230537244672,
      "utilisation": 0.9005,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 270296690688,
      "utilisation": 0.9385,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 270296690688,
      "utilisation": 0.9385,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 403889922048,
      "utilisation": 0.9349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.9203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 7.0088,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 140176625664,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.1903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.4602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 189376585728,
      "utilisation": 0.9863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 3.8938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 403889922048,
      "utilisation": 0.7888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.1903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.1903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 403889922048,
      "utilisation": 0.7888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 14.0177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.7522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.7522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 140176625664,
      "utilisation": 0.7788,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 262651097088,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.0951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 12.7433,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 35.0442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 12.7433,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 35.0442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 14.0177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 23.3628,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 17.5221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 11.6814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 8.761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 4.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.7522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.7522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 140176625664,
      "utilisation": 0.9942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 140176625664,
      "utilisation": 0.9942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 5.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.9203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.9203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.9203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 2.9203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 140176625664,
      "utilisation": 1.4602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44176625664
    },
    {
      "model_slug": "ifm-k2-horizon-375b-a23b",
      "model_name": "K2-Horizon-375B-A23B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-375B-A23B",
      "config_revision": "236f3183898840c50f10ab9bd948f60d2d5e4381",
      "parameters": 379167159168,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 270296690688,
      "utilisation": 0.9385,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0463,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.3128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.556,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.3128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.3128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9401335808,
      "utilisation": 0.9401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0741,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.1564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9401335808,
      "utilisation": 0.8547,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5676818432,
      "utilisation": 1.4192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1676818432
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9401335808,
      "utilisation": 0.8547,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5676818432,
      "utilisation": 1.4192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1676818432
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9401335808,
      "utilisation": 0.9401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5676818432,
      "utilisation": 0.9461,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7599005696,
      "utilisation": 0.9499,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11580751872,
      "utilisation": 0.7238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.6255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.142,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.142,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.2085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ifm-k2-horizon-7b",
      "model_name": "K2-Horizon-7B",
      "publisher": "IFM",
      "hf_repo": "IFM/K2-Horizon-7B",
      "config_revision": "30d38fecf8a609873ae73a617f5c714286e1f565",
      "parameters": 8999178240,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20017201152,
      "utilisation": 0.0695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0935,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0701,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0623,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0623,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.3738,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.8971,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1121,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2804,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1869,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0935,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.4984,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.035,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2804,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2804,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.035,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2243,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2243,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0997,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.9468,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5052873024,
      "utilisation": 1.2632,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1052873024
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.9468,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5052873024,
      "utilisation": 1.2632,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1052873024
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512448,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.5607,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2243,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2243,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1273,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1273,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.7476,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.3738,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.3738,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.3738,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.3738,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.2492,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.1869,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ilyagusev-saiga-llama3-8b",
      "model_name": "saiga_llama3_8b",
      "publisher": "IlyaGusev",
      "hf_repo": "IlyaGusev/saiga_llama3_8b",
      "config_revision": "5bb9917bdb85340549662ebb62c8e522037ff3f3",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17942632768,
      "utilisation": 0.0623,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.7126,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.5345,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.8899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.8899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.5933,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.0767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.584,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.8551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 51680978346,
      "utilisation": 0.8075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 91662339488,
      "utilisation": 0.9548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.7126,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.4356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.5006,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 51680978346,
      "utilisation": 0.8075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 51680978346,
      "utilisation": 0.8075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.5006,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 5.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78206506400,
      "utilisation": 0.9776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78206506400,
      "utilisation": 0.9776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.7601,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.9492,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 105963291040,
      "utilisation": 0.8278,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.6983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 12.9202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.6983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 12.9202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 5.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 8.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 6.4601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 4.3067,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 3.2301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78206506400,
      "utilisation": 0.9776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78206506400,
      "utilisation": 0.9776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.9704,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136821312000,
      "utilisation": 0.9704,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 2.1534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.0767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.0767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.0767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 51680978346,
      "utilisation": 1.0767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3680978346
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 64955500329,
      "utilisation": 0.9022,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 91662339488,
      "utilisation": 0.9548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-flash",
      "model_name": "Ling-3.0-flash",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-flash",
      "config_revision": "42766a814ab117e75e2e61465d5e131b72d931a3",
      "parameters": 127486405600,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 256287445728,
      "utilisation": 0.8899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0877,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0657,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.039,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.3506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.8415,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.263,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1753,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0877,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.4675,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.263,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.263,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0329,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.9435,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.2104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.2104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0935,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0623,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1315,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.8577,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4152275965,
      "utilisation": 1.0381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152275965
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.8577,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4152275965,
      "utilisation": 1.0381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152275965
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.9435,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5850387360,
      "utilisation": 0.9751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7525754784,
      "utilisation": 0.9407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.7862,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9434680224,
      "utilisation": 0.5897,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.5259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.2104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.2104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1194,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1194,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.7012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.3506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.3506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.3506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.3506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.2337,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.1753,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ling-3-0-tiny",
      "model_name": "Ling-3.0-tiny",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ling-3.0-tiny",
      "config_revision": "b61f4338de3e68ffc9c0bc1ed5e902981a4a929e",
      "parameters": 7893392800,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16829967680,
      "utilisation": 0.0584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.0777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.5243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.3495,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.9321,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.0655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.5243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.5243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.0655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9249036288,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.4194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.4194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1864,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.2621,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.7932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.7932,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9249036288,
      "utilisation": 0.9249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7172880384,
      "utilisation": 1.1955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1172880384
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7172880384,
      "utilisation": 0.8966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10998052864,
      "utilisation": 0.9165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14459705344,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.4194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.4194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18379708416,
      "utilisation": 0.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.3495,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-llada2-0-mini",
      "model_name": "LLaDA2.0-mini",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/LLaDA2.0-mini",
      "config_revision": "dad945cac317da394b390f82c7b40691d8a881ed",
      "parameters": 16255643392,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 33553913856,
      "utilisation": 0.1165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.8916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.4187,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.2611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.2611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 363192907776,
      "utilisation": 0.8407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 7.5665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 18.1596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 343192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.27,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 203192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 5.6749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 299192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 3.7833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 267192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.8916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 10.0887,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 327192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 494059419648,
      "utilisation": 0.965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 5.6749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 299192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 5.6749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 299192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 494059419648,
      "utilisation": 0.965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 36.3193,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 4.5399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 283192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 4.5399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 283192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.0177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 183192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.3452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 93192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 33.0175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 352192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 90.7982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 359192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 33.0175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 352192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 90.7982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 359192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 36.3193,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 60.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 45.3991,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 355192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 30.2661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 351192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 22.6996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 11.3498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 331192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 4.5399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 283192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 4.5399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 283192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.5758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 2.5758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 15.133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 339192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 7.5665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 7.5665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 7.5665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 7.5665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 5.0443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 291192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 3.7833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 267192907776
    },
    {
      "model_slug": "inclusionai-ring-2-5-1t",
      "model_name": "Ring-2.5-1T",
      "publisher": "inclusionAI",
      "hf_repo": "inclusionAI/Ring-2.5-1T",
      "config_revision": "235b2c183a5e1c2fe610858d33b81d4516230db0",
      "parameters": 1012474606720,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 363192907776,
      "utilisation": 1.2611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75192907776
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 418807304192,
      "utilisation": 0.9695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 13.5107,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 250214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.6888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 7.506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 234214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.5012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 198214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "internlm-atria-dawn-preview",
      "model_name": "Atria-Dawn-Preview",
      "publisher": "internlm",
      "hf_repo": "internlm/Atria-Dawn-Preview",
      "config_revision": "311a9b8c6c0e8dd4b00b2d7baff7a926c3415e78",
      "parameters": 753329940480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18592598246,
      "utilisation": 0.9296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.4455,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.8296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.99,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1",
      "model_name": "Agents-A1",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1",
      "config_revision": "addff08f1653ee72765c5cf458fe84556bb34f8e",
      "parameters": 35107181936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.5102,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2835,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.9277,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3908454463,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.9277,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3908454463,
      "utilisation": 0.9771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.9915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5948854321,
      "utilisation": 0.7436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.8504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1417,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.1063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "internscience-agents-a1-4b",
      "model_name": "Agents-A1-4B",
      "publisher": "InternScience",
      "hf_repo": "InternScience/Agents-A1-4B",
      "config_revision": "945c40a4aa6f534d434a353207b8d42ecf7a5293",
      "parameters": 4539265536,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10204415761,
      "utilisation": 0.0354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "jica98-qwen3-5-4b-super-coder",
      "model_name": "qwen3.5-4B-super-coder",
      "publisher": "jica98",
      "hf_repo": "jica98/qwen3.5-4B-super-coder",
      "config_revision": "1f2362bff662708edca4836deda762da0c046dd0",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18098272256,
      "utilisation": 0.9049,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.8086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.9632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "kwaipilot-kat-coder-v2-5-dev",
      "model_name": "KAT-Coder-V2.5-Dev",
      "publisher": "Kwaipilot",
      "hf_repo": "Kwaipilot/KAT-Coder-V2.5-Dev",
      "config_revision": "7be56fe773e72b6f5ca93c1ae45d828ddb893922",
      "parameters": 34660610688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0912,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0684,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0608,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0608,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.3649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.8758,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1095,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2737,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1825,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0912,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.4865,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0342,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2737,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2737,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0342,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8292359808,
      "utilisation": 0.8292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0973,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4855416832,
      "utilisation": 1.2139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 855416832
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4855416832,
      "utilisation": 1.2139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 855416832
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8292359808,
      "utilisation": 0.8292,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5795596288,
      "utilisation": 0.9659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7443406464,
      "utilisation": 0.9304,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.8488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10185825920,
      "utilisation": 0.6366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.5474,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1242,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1242,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.7298,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.3649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.3649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.3649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.3649,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.2433,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.1825,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-3-5-7-8b-instruct",
      "model_name": "EXAONE-3.5-7.8B-Instruct",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
      "config_revision": "553ea250b9a5317231459279d5847d6cf955b9aa",
      "parameters": 7818448896,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17515372160,
      "utilisation": 0.0608,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0893,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0446,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.4286,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0335,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2890213376,
      "utilisation": 0.7226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2890213376,
      "utilisation": 0.7226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.4286,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.7144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.5358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.3572,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.2679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.1786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0893,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0893,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0893,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0893,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0446,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-1-2b",
      "model_name": "EXAONE-4.0-1.2B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-1.2B",
      "config_revision": "3abf2810673c7c0778df64a73c2d52eab32d91c4",
      "parameters": 1279391488,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4286130176,
      "utilisation": 0.0149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.3446,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.2584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.2297,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.2297,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.1531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.7532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 17923692544,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.4135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.5649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.6891,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.3446,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.7889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.1292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.5649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.5649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.1292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.3854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.8269,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.8269,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.3675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.245,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.5168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.2594,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 3.4634,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.2594,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 3.4634,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.3854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 2.309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.7317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13853743104,
      "utilisation": 1.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1853743104
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13853743104,
      "utilisation": 0.8659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28401260544,
      "utilisation": 0.8875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.8269,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.8269,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.4692,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.4692,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21486065664,
      "utilisation": 0.8953,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.7532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.7532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.7532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36151875584,
      "utilisation": 0.7532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.9188,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.6891,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "lgai-exaone-exaone-4-0-32b",
      "model_name": "EXAONE-4.0-32B",
      "publisher": "LGAI-EXAONE",
      "hf_repo": "LGAI-EXAONE/EXAONE-4.0-32B",
      "config_revision": "a1d54d1c148c30881ed27e035b650da489b51b92",
      "parameters": 32003216384,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 66154256384,
      "utilisation": 0.2297,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0169,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0127,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0507,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0169,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0507,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0507,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.3243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0253,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.8107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.8107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.3243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.5404,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.4053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.2027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.1351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-1-2b-instruct",
      "model_name": "LFM2.5-1.2B-Instruct",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-1.2B-Instruct",
      "config_revision": "0f604ada3f766f9f257460c4c9f0b5d6f69d431b",
      "parameters": 1170340608,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3242614320,
      "utilisation": 0.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.033,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0247,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.022,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.022,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0147,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3166,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.066,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.033,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1759,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.6332,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.9524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.9524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.6332,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3809538048,
      "utilisation": 0.6349,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.7914,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.5276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.3957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1979,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.2638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.1319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.0879,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.066,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-2-6b",
      "model_name": "LFM2.5-2.6B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-2.6B",
      "config_revision": "654f9463ce32b05d0429d76fe1f580b27d4c1ac0",
      "parameters": 2697198592,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6331515633,
      "utilisation": 0.022,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0031,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.068,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0027,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0106,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.3401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.3401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.136,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.2267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.17,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.1134,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0096,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0096,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-230m",
      "model_name": "LFM2.5-230M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-230M",
      "config_revision": "40cb2ad3b3044d5a41eee083a6103c8b523afa45",
      "parameters": 229693184,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1360348950,
      "utilisation": 0.0047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0037,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0335,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0168,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0447,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0031,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0031,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.4025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.4025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.2013,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.1006,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0335,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0335,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0335,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0335,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0168,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-350m",
      "model_name": "LFM2.5-350M",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-350M",
      "config_revision": "9e6c6ccf47cd318696e137d381a7ded8fe4df09f",
      "parameters": 354483968,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1610064285,
      "utilisation": 0.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.9071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2835,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2835,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2835,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.9911,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.0672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4136454144,
      "utilisation": 1.0341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136454144
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4136454144,
      "utilisation": 1.0341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136454144
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.9911,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5745784032,
      "utilisation": 0.9576,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7860892896,
      "utilisation": 0.9826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.8259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9911301344,
      "utilisation": 0.6195,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.2268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.1287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.252,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-8b-a1b",
      "model_name": "LFM2.5-8B-A1B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-8B-A1B",
      "config_revision": "5dd22602c2e9f6a097b1de4c4efe0658b605015c",
      "parameters": 8467856832,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18142857216,
      "utilisation": 0.063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0281,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0166,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.3592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1123,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.014,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1123,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1123,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.014,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.7184,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0898,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0898,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.6531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3499919543,
      "utilisation": 0.875,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.6531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3499919543,
      "utilisation": 0.875,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.7184,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4256193079,
      "utilisation": 0.7094,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.8981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.5987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0898,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0898,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.051,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.2994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.1497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "liquidai-lfm2-5-vl-3b",
      "model_name": "LFM2.5-VL-3B",
      "publisher": "LiquidAI",
      "hf_repo": "LiquidAI/LFM2.5-VL-3B",
      "config_revision": "a3af5799199acdd2a4f56ac4342816abb46c12a9",
      "parameters": 3123483888,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7184459224,
      "utilisation": 0.0249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.3419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.2564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.2279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.2279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.152,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19589607424,
      "utilisation": 0.9795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.4103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.6838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.3419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.9868,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.1282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.1282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.3204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.8206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.8206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.3647,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.2431,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.5128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.2004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 3.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.2004,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 3.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.3204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 2.2007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.6505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13204107264,
      "utilisation": 1.1003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1204107264
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13204107264,
      "utilisation": 0.8253,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27742941184,
      "utilisation": 0.867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.8206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.8206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.4656,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.4656,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 23542738944,
      "utilisation": 0.9809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35524030464,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.9117,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.6838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-32b-a3b-thinking",
      "model_name": "llm-jp-4.1-32b-a3b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-32b-a3b-thinking",
      "config_revision": "cda260706786758045e5e96bf4d738bbc01155b5",
      "parameters": 32139028992,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65644376064,
      "utilisation": 0.2279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.2711,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.1606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.7969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19460216832,
      "utilisation": 0.973,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.4337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.5977,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.7229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.8391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.5977,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.5977,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.5308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.8675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.8675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.3855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.5422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.3916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 3.827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.3916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 3.827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.5308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 2.5513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.9135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15307855872,
      "utilisation": 1.2757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3307855872
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15307855872,
      "utilisation": 0.9567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30208120832,
      "utilisation": 0.944,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.8675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.8675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.4922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.4922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23114504192,
      "utilisation": 0.9631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.7969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.7969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.7969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38253320192,
      "utilisation": 0.7969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.9638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.7229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-33b-thinking",
      "model_name": "llm-jp-4.1-33b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-33b-thinking",
      "config_revision": "de581ee4353568c480d47129948d1be99f02f4ff",
      "parameters": 33219548160,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69396027392,
      "utilisation": 0.241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0745,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.3971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.9531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2979,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0372,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2979,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2979,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0372,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8929363968,
      "utilisation": 0.8929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8929363968,
      "utilisation": 0.8118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5302339584,
      "utilisation": 1.3256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1302339584
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8929363968,
      "utilisation": 0.8118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5302339584,
      "utilisation": 1.3256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1302339584
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8929363968,
      "utilisation": 0.8929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5302339584,
      "utilisation": 0.8837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7895468032,
      "utilisation": 0.9869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.9175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11009738752,
      "utilisation": 0.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.5957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.3971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.3971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.3971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.3971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.2648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.1986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "llm-jp-llm-jp-4-1-8b-thinking",
      "model_name": "llm-jp-4.1-8b-thinking",
      "publisher": "llm-jp",
      "hf_repo": "llm-jp/llm-jp-4.1-8b-thinking",
      "config_revision": "7c19bc90e49b717381e4f474ee95964c05217a28",
      "parameters": 8590200832,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19062802432,
      "utilisation": 0.0662,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 3.6685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 512360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 2.7514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 448360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 2.4457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 416360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 2.4457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 416360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 1.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 14.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 656360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 35.218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 684360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 4.4023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 544360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 11.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 640360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 7.3371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 608360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 3.6685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 512360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 19.5656,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 668360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 1.3757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 11.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 640360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 11.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 640360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 1.3757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 70.436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 694360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 8.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 624360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 8.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 624360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 3.9131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 524360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 2.6087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 434360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 5.5028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 576360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 64.0328,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 693360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 176.0901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 700360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 64.0328,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 693360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 176.0901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 700360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 70.436,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 694360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 117.3934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 698360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 88.045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 58.6967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 692360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 44.0225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 688360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 22.0113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 672360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 8.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 624360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 8.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 624360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 4.9955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 563360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 4.9955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 563360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 29.3483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 680360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 14.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 656360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 14.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 656360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 14.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 656360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 14.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 656360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 9.7828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 632360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 7.3371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 608360259258
    },
    {
      "model_slug": "meituan-longcat-longcat-2-0",
      "model_name": "LongCat-2.0",
      "publisher": "meituan-longcat",
      "hf_repo": "meituan-longcat/LongCat-2.0",
      "config_revision": "834bf5ffe3047aa9f6cc7a64a9bc068b146b8274",
      "parameters": 1775560491136,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 704360259258,
      "utilisation": 2.4457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 416360259258
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.3155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.2366,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.2103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.2103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.1402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.6803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 19269967540,
      "utilisation": 0.9635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.3786,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.9464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.6309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.3155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.9071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.1183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.9464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.9464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.1183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.2782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.7571,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.7571,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.3365,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.2243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 3.1956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 3.1956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.2782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 2.1304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.5978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12782385000,
      "utilisation": 1.0652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 782385000
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15882876254,
      "utilisation": 0.9927,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25444895500,
      "utilisation": 0.7952,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.7571,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.7571,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.4296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22262518522,
      "utilisation": 0.9276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.6803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.6803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.6803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32654561236,
      "utilisation": 0.6803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.8413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.6309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "meta-models-muse-glimmer-30b",
      "model_name": "Muse-Glimmer-30B",
      "publisher": "meta-models",
      "hf_repo": "meta-models/Muse-Glimmer-30B",
      "config_revision": "a4e59da52a7bc87ae7251dd5545c0dd437c44b68",
      "parameters": 29776626688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 60570148756,
      "utilisation": 0.2103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-2",
      "model_name": "phi-2",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-2",
      "config_revision": "810d367871c1d460086d9f82db8696f2e0a0fcd0",
      "parameters": 2779683840,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0456,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.5833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0729,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.8083,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0432,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.7348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5438235648,
      "utilisation": 1.3596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1438235648
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.7348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5438235648,
      "utilisation": 1.3596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1438235648
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.8083,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-5-mini-instruct",
      "model_name": "Phi-3.5-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3.5-mini-instruct",
      "config_revision": "2fe192450127e6a83f7441aef6e3ca586c338b77",
      "parameters": 3821079552,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0456,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.027,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.5833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0729,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0608,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.8083,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0432,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.7348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5438235648,
      "utilisation": 1.3596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1438235648
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.7348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5438235648,
      "utilisation": 1.3596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1438235648
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8083253760,
      "utilisation": 0.8083,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5920098816,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7157884416,
      "utilisation": 0.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.7291,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1458,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.4861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.1215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-128k-instruct",
      "model_name": "Phi-3-mini-128k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-128k-instruct",
      "config_revision": "f3c06aed622e14ca0abf5115094e4fc9a9948f36",
      "parameters": 3821079552,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 11665328640,
      "utilisation": 0.0405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-3-mini-4k-instruct",
      "model_name": "Phi-3-mini-4k-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-3-mini-4k-instruct",
      "config_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
      "parameters": 3821079552,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1242,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.0736,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.6625,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.9029,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1988,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.4969,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3313,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.8834,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.0621,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.4969,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.4969,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.0621,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9751189504,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1767,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1178,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2484,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10861140160,
      "utilisation": 0.9874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 2.0063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10861140160,
      "utilisation": 0.9874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 2.0063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9751189504,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.3375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2025069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8025069760,
      "utilisation": 1.0031,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25069760
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11368740864,
      "utilisation": 0.9474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14507972800,
      "utilisation": 0.9067,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.9938,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2255,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.2255,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18058221760,
      "utilisation": 0.7524,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.6625,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.6625,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.6625,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.6625,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.4417,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.3313,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4",
      "model_name": "phi-4",
      "publisher": "microsoft",
      "hf_repo": "microsoft/phi-4",
      "config_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
      "parameters": 14659507200,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31801120960,
      "utilisation": 0.1104,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0562,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.5392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1685,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0562,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0211,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1685,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1685,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0211,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.6611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.9803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3737608192,
      "utilisation": 0.9344,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.9803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3737608192,
      "utilisation": 0.9344,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.6611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5533539328,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6611375104,
      "utilisation": 0.8264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.4493,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.1123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-4-mini-instruct",
      "model_name": "Phi-4-mini-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-4-mini-instruct",
      "config_revision": "cfbefacb99257ffa30c83adab238a50856ac3083",
      "parameters": 3836021760,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10783642624,
      "utilisation": 0.0374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "microsoft-phi-tiny-moe-instruct",
      "model_name": "Phi-tiny-MoE-instruct",
      "publisher": "microsoft",
      "hf_repo": "microsoft/Phi-tiny-MoE-instruct",
      "config_revision": "2fe50e88d0e2a5a132563815686ea0dcc8e252b5",
      "parameters": 3755220288,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.5695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 4.3018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 141220894720,
      "utilisation": 0.8826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.3899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 50036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 165182445568,
      "utilisation": 0.9177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.9112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2",
      "model_name": "MiniMax-M2",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2",
      "config_revision": "757303d492a50514c312788b5247a4f696a4c6a3",
      "parameters": 228689764864,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.5695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 4.3018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 141220894720,
      "utilisation": 0.8826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.3899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 50036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 165182445568,
      "utilisation": 0.9177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.9112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-1",
      "model_name": "MiniMax-M2.1",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.1",
      "config_revision": "cd97f59135f37b2a6bf09356e485d5e4aeb7dc9c",
      "parameters": 228689764864,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.5695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 4.3018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 141220894720,
      "utilisation": 0.8826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.3899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 50036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 165182445568,
      "utilisation": 0.9177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.9112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-5",
      "model_name": "MiniMax-M2.5",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.5",
      "config_revision": "f710177d938eff80b684d42c5aa84b382612f21f",
      "parameters": 228703644928,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.5695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 4.3018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 141220894720,
      "utilisation": 0.8826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 190641593344,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.3899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 50036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.3443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 460365534208,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 165182445568,
      "utilisation": 0.9177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.9112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 116083446784,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.8215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 21.5091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 8.6036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 14.3394,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 10.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 7.1697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 5.3773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 2.6886,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.0755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 131853752320,
      "utilisation": 0.9351,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 3.5849,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.7924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 86036439040,
      "utilisation": 1.195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14036439040
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 86036439040,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m2-7",
      "model_name": "MiniMax-M2.7",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M2.7",
      "config_revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b",
      "parameters": 228689764864,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 246015382528,
      "utilisation": 0.8542,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 170914528463,
      "utilisation": 0.8902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 250290614516,
      "utilisation": 0.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 263955899001,
      "utilisation": 0.9165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 263955899001,
      "utilisation": 0.9165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 352513348066,
      "utilisation": 0.816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 3.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 8.5457,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.0682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.6705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.7804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 170914528463,
      "utilisation": 0.8902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 4.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 455910442003,
      "utilisation": 0.8905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.6705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.6705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 455910442003,
      "utilisation": 0.8905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 17.0915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.1364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.1364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 170914528463,
      "utilisation": 0.9495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 263955899001,
      "utilisation": 0.9776,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.3353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 15.5377,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 159914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 42.7286,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 15.5377,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 159914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 42.7286,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 17.0915,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 28.4858,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 21.3643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 162914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 14.2429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 158914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 10.6822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 5.3411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.1364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.1364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.2122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.2122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 7.1214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 3.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 3.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 3.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 3.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 2.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 170914528463,
      "utilisation": 1.7804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74914528463
    },
    {
      "model_slug": "minimaxai-minimax-m3",
      "model_name": "MiniMax-M3",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-M3",
      "config_revision": "f0e1c1e04d40177e4673a22097036854f536e9c0",
      "parameters": 427040140160,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 263955899001,
      "utilisation": 0.9165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 184095858057,
      "utilisation": 0.9588,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 231586193415,
      "utilisation": 0.9046,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 283466391705,
      "utilisation": 0.9843,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 283466391705,
      "utilisation": 0.9843,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 378047983972,
      "utilisation": 0.8751,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 3.8353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 9.2048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 164095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.1506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.8765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.9177,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 184095858057,
      "utilisation": 0.9588,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.1138,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 488478691760,
      "utilisation": 0.9541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.8765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.8765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 488478691760,
      "utilisation": 0.9541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 18.4096,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.3012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.3012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.0228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 268871522735,
      "utilisation": 0.9958,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.4382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 16.736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 173095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 46.024,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 16.736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 173095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 46.024,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 18.4096,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 30.6826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 178095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 23.012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 176095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 15.3413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 172095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 11.506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 168095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 5.753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.3012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.3012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.3056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.3056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 7.6707,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 160095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 3.8353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 3.8353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 3.8353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 3.8353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 2.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 184095858057,
      "utilisation": 1.9177,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88095858057
    },
    {
      "model_slug": "minimaxai-minimax-text-01",
      "model_name": "MiniMax-Text-01",
      "publisher": "MiniMaxAI",
      "hf_repo": "MiniMaxAI/MiniMax-Text-01",
      "config_revision": "a7351bf2bee0e1253919d349f1ad304e6dac13e9",
      "parameters": 456089655296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 283466391705,
      "utilisation": 0.9843,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.7115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.9915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.8813,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.8813,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.5875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.0176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.4422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.8538,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 48844490752,
      "utilisation": 0.7632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 92067186688,
      "utilisation": 0.959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.7115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.3568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.4957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 48844490752,
      "utilisation": 0.7632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 48844490752,
      "utilisation": 0.7632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.4957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.8844,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78647511040,
      "utilisation": 0.9831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78647511040,
      "utilisation": 0.9831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.7589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.9401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 106325592064,
      "utilisation": 0.8307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.4404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 12.2111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.4404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 12.2111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.8844,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 8.1407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 6.1056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 4.0704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 3.0528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.5264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78647511040,
      "utilisation": 0.9831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 78647511040,
      "utilisation": 0.9831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.9688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 136604796928,
      "utilisation": 0.9688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 2.0352,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.0176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.0176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.0176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 48844490752,
      "utilisation": 1.0176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 844490752
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 64389105664,
      "utilisation": 0.8943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 92067186688,
      "utilisation": 0.959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-2-123b-instruct-2512",
      "model_name": "Devstral-2-123B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-2-123B-Instruct-2512",
      "config_revision": "1613bf01adb5e1c6fdc196b46e6b173eae75eb4a",
      "parameters": 125025989840,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 253814622208,
      "utilisation": 0.8813,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.2567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1141,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18906168064,
      "utilisation": 0.9453,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3081,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.7702,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.5135,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.2567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.7555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.7702,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.7702,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.2739,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3496,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.3496,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27196963584,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.6847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.5135,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-devstral-small-2507",
      "model_name": "Devstral-Small-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Devstral-Small-2507",
      "config_revision": "bd165ab26cebbcc2eea2c4ecbfc07f3ac42b3c39",
      "parameters": 23572403200,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49295702784,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.2568,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1926,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1141,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18906684416,
      "utilisation": 0.9453,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3081,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.5135,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.2568,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.7555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.7703,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.0963,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.2739,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1826,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3851,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.8499,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.6162,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3496,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.3496,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27197479936,
      "utilisation": 0.5666,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.6847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.5135,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-magistral-small-2509",
      "model_name": "Magistral-Small-2509",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Magistral-Small-2509",
      "config_revision": "a31cc96ab10cf19bc42c628fedf1e359e0853c49",
      "parameters": 24011361280,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49296219136,
      "utilisation": 0.1712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1174,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.0695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.6259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.8251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.4694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.0587,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.4694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.4694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.0587,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9116436529,
      "utilisation": 0.9116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.3756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.3756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1669,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10381770304,
      "utilisation": 0.9438,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.9161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 10381770304,
      "utilisation": 0.9438,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.9161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9116436529,
      "utilisation": 0.9116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7664410047,
      "utilisation": 1.2774,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1664410047
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7664410047,
      "utilisation": 0.9581,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 11763269184,
      "utilisation": 0.9803,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 13594535007,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.9389,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.3756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.3756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.2131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 16502013504,
      "utilisation": 0.6876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.6259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.6259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.6259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.6259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-14b-instruct-2512",
      "model_name": "Ministral-3-14B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-14B-Instruct-2512",
      "config_revision": "29439f81c2be264d8d393273f99e7db9c0961120",
      "parameters": 13945032240,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 30044443663,
      "utilisation": 0.1043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0216,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.4671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0973,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0182,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0182,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.9342,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.8492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3819438240,
      "utilisation": 0.9549,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.8492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3819438240,
      "utilisation": 0.9549,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.9342,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.8874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5324619936,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.7785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.5838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.2919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1168,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0663,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0663,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.3892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.1297,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0973,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-3b-instruct-2512",
      "model_name": "Ministral-3-3B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-3B-Instruct-2512",
      "config_revision": "b35d4dfe56c142746f54dbd64f579faab2744308",
      "parameters": 3849090048,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9341564928,
      "utilisation": 0.0324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0773,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0458,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.4122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.9892,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.3091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2061,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.5496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.3091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.3091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9264780129,
      "utilisation": 0.9265,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0733,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9973,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5472389268,
      "utilisation": 1.3681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1472389268
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9973,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5472389268,
      "utilisation": 1.3681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1472389268
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9264780129,
      "utilisation": 0.9265,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5472389268,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7139762592,
      "utilisation": 0.8925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.9142,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10970243488,
      "utilisation": 0.6856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.6183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.1403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.8244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.4122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.4122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.4122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.4122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.2061,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-ministral-3-8b-instruct-2512",
      "model_name": "Ministral-3-8B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Ministral-3-8B-Instruct-2512",
      "config_revision": "5b26027e7b19eeb4b7352e1fed3926375dd2cb4d",
      "parameters": 8918026716,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19784707393,
      "utilisation": 0.0687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.3408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.818,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2556,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2556,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2556,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.87,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4521439232,
      "utilisation": 1.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 521439232
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.87,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4521439232,
      "utilisation": 1.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 521439232
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5983465472,
      "utilisation": 0.9972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7816613888,
      "utilisation": 0.9771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.7975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9570406400,
      "utilisation": 0.5982,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.5112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.6816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.3408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.3408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.3408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.3408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.2272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.1704,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-2",
      "model_name": "Mistral-7B-Instruct-v0.2",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.2",
      "config_revision": "63a8b081895390a26e140280378bc85ec8bce07a",
      "parameters": 7241732096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16359280640,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.064,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.3411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.8186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2558,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1705,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.4548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2558,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2558,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.9577,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.091,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.8706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4525082624,
      "utilisation": 1.1313,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 525082624
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.8706,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4525082624,
      "utilisation": 1.1313,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 525082624
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.9577,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 5987846144,
      "utilisation": 0.998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7821805568,
      "utilisation": 0.9777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.7981,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9577121792,
      "utilisation": 0.5986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.5116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.6822,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.3411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.3411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.3411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.3411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.2274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.1705,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-instruct-v0-3",
      "model_name": "Mistral-7B-Instruct-v0.3",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-Instruct-v0.3",
      "config_revision": "c170c708c41dac9275d15a8fff4eca08d52bab71",
      "parameters": 7248023552,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16371894272,
      "utilisation": 0.0568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.3296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.7911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0989,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.2472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4395,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.2472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.2472,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.9034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3984568320,
      "utilisation": 0.9961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3984568320,
      "utilisation": 0.9961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.9034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5706117120,
      "utilisation": 0.951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7279742976,
      "utilisation": 0.91,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9033535488,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.9889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.4945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1978,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.3296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.3296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.3296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.3296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.2198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.1648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-7b-v0-1",
      "model_name": "Mistral-7B-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-7B-v0.1",
      "config_revision": "27d67f1b5f57dc0953326b2601d68371d40ea8da",
      "parameters": 7241732096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15822409728,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.2688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 243609965568,
      "utilisation": 0.9516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 243609965568,
      "utilisation": 0.8459,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 243609965568,
      "utilisation": 0.8459,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 404115990528,
      "utilisation": 0.9355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 5.0752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 12.1805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 223609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.5226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.8064,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 179609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 2.5376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 147609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.2688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 6.7669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 474740295680,
      "utilisation": 0.9272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.8064,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 179609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.8064,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 179609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 474740295680,
      "utilisation": 0.9272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 24.361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 233609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 243609965568,
      "utilisation": 0.9023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 22.1464,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 60.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 239609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 22.1464,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 60.9025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 239609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 24.361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 233609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 40.6017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 30.4512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 235609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 20.3008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 15.2256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 227609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 7.6128,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 163609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.7277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 1.7277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 10.1504,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 5.0752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 5.0752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 5.0752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 5.0752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 195609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 3.3835,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 243609965568,
      "utilisation": 2.5376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 147609965568
    },
    {
      "model_slug": "mistralai-mistral-large-3-675b-instruct-2512",
      "model_name": "Mistral-Large-3-675B-Instruct-2512",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Large-3-675B-Instruct-2512",
      "config_revision": "383ffea2c7d60dfd44ca960e8e691709d4fdb9cd",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 243609965568,
      "utilisation": 0.8459,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.1388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.1041,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.0925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.0925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.7581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.1665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.4163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2775,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.1388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.7401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.052,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.4163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.4163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.052,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9617004544,
      "utilisation": 0.9617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.3331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.3331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.148,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.0987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.7044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.7044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9617004544,
      "utilisation": 0.9617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6817798144,
      "utilisation": 1.1363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817798144
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6817798144,
      "utilisation": 0.8522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10867431424,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.9476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.8326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.3331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.3331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15162169344,
      "utilisation": 0.6318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.5551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.3701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.2775,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-nemo-instruct-2407",
      "model_name": "Mistral-Nemo-Instruct-2407",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Nemo-Instruct-2407",
      "config_revision": "04d8a90549d23fc6bd7f642064003592df51e9b3",
      "parameters": 12247782400,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26644076544,
      "utilisation": 0.0925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1926,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18903783424,
      "utilisation": 0.9452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.7554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6846,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-24b-instruct-2501",
      "model_name": "Mistral-Small-24B-Instruct-2501",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-24B-Instruct-2501",
      "config_revision": "9527884be6e5616bdd54de542f9ae13384489724",
      "parameters": 23572403200,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1926,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18903783424,
      "utilisation": 0.9452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.7554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6846,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-1-24b-instruct-2503",
      "model_name": "Mistral-Small-3.1-24B-Instruct-2503",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.1-24B-Instruct-2503",
      "config_revision": "68faf511d618ef198fef186659617cfd2eb8e33a",
      "parameters": 24011361280,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1926,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 18903783424,
      "utilisation": 0.9452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.7554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.7702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.0963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.2739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 2.7293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.8195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 10917074944,
      "utilisation": 1.3646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2917074944
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 10917074944,
      "utilisation": 0.9098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 15581599744,
      "utilisation": 0.9738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.8498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.3496,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 21485737984,
      "utilisation": 0.8952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27194578944,
      "utilisation": 0.5666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.6846,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.5135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-3-2-24b-instruct-2506",
      "model_name": "Mistral-Small-3.2-24B-Instruct-2506",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
      "config_revision": "95a6d26c4bfb886c58daf9d3f7332c857cb27b43",
      "parameters": 24011361280,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 49293318144,
      "utilisation": 0.1712,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.6588,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.9264,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.8235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.8235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 44001368064,
      "utilisation": 0.9167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.2001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.7906,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59565242368,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84870854656,
      "utilisation": 0.8841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.6588,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.2223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.4632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59565242368,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59565242368,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.4632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 4.4001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 72601466880,
      "utilisation": 0.9075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 72601466880,
      "utilisation": 0.9075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.7028,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.8784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.9883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 4.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 11.0003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 4.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 11.0003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 4.4001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 7.3336,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 5.5002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 3.6668,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 2.7501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 72601466880,
      "utilisation": 0.9075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 72601466880,
      "utilisation": 0.9075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.8971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 126496741376,
      "utilisation": 0.8971,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 44001368064,
      "utilisation": 1.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001368064
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 44001368064,
      "utilisation": 0.9167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 44001368064,
      "utilisation": 0.9167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 44001368064,
      "utilisation": 0.9167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 44001368064,
      "utilisation": 0.9167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67611334656,
      "utilisation": 0.939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84870854656,
      "utilisation": 0.8841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mistral-small-4-119b-2603",
      "model_name": "Mistral-Small-4-119B-2603",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mistral-Small-4-119B-2603",
      "config_revision": "a11f36bebf709121056b1dbcc943d1c6afbe494d",
      "parameters": 119401317952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 237166401536,
      "utilisation": 0.8235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.7922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.5942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.9859,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.9859,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.6573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.1157,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.6776,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.9507,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 53552572416,
      "utilisation": 0.8368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 87707510784,
      "utilisation": 0.9136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.7922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.4876,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.5546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 53552572416,
      "utilisation": 0.8368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 53552572416,
      "utilisation": 0.8368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.5546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 5.3553,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 72075405312,
      "utilisation": 0.9009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 72075405312,
      "utilisation": 0.9009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.8451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 152110196736,
      "utilisation": 0.5634,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.9223,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.8684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 13.3881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.8684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 13.3881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 5.3553,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 8.9254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 6.6941,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 4.4627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 3.347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.6735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 72075405312,
      "utilisation": 0.9009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 72075405312,
      "utilisation": 0.9009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.8372,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 118052186112,
      "utilisation": 0.8372,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 2.2314,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.1157,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.1157,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.1157,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 53552572416,
      "utilisation": 1.1157,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5552572416
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 53552572416,
      "utilisation": 0.7438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 87707510784,
      "utilisation": 0.9136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x22b-instruct-v0-1",
      "model_name": "Mixtral-8x22B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x22B-Instruct-v0.1",
      "config_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
      "parameters": 140630071296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 283947657216,
      "utilisation": 0.9859,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.4963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.3722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.3308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.3308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.2206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40190349312,
      "utilisation": 0.8373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.5955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.8047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.4963,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 35012374528,
      "utilisation": 0.9726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.1861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.8047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.8047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.1861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.8809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.6438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.6438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.5294,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.3529,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.7444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.7099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 4.7023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.7099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 4.7023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.8809,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 3.1349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 2.3512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.5674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18809335808,
      "utilisation": 1.1756,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2809335808
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30138986496,
      "utilisation": 0.9418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.6438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.6438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.6758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.6758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18809335808,
      "utilisation": 0.7837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40190349312,
      "utilisation": 0.8373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40190349312,
      "utilisation": 0.8373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40190349312,
      "utilisation": 0.8373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40190349312,
      "utilisation": 0.8373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 51500863488,
      "utilisation": 0.7153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.9925,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-mixtral-8x7b-instruct-v0-1",
      "model_name": "Mixtral-8x7B-Instruct-v0.1",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Mixtral-8x7B-Instruct-v0.1",
      "config_revision": "eba92302a2861cdc0098cc54bc9f17cb2c47eb61",
      "parameters": 46702792704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 95283499008,
      "utilisation": 0.3308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.1173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.5821,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 19474201341,
      "utilisation": 0.9737,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.3168,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.528,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.7762,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.6336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.6336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.2816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.1877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.0682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 2.9375,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.0682,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 2.9375,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.9583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 11749850460,
      "utilisation": 1.4687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3749850460
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 11749850460,
      "utilisation": 0.9792,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14276110485,
      "utilisation": 0.8923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.8732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.6336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.6336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.3595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.3595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 22067181318,
      "utilisation": 0.9195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.5821,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.5821,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.5821,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 27941569876,
      "utilisation": 0.5821,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.704,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.528,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "mistralai-voxtral-small-24b-2507",
      "model_name": "Voxtral-Small-24B-2507",
      "publisher": "mistralai",
      "hf_repo": "mistralai/Voxtral-Small-24B-2507",
      "config_revision": "da5b42409f279fdd92febee0511a6c32828569c1",
      "parameters": 24261800960,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 50687008276,
      "utilisation": 0.176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.5522,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 397375668224,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 19.8688,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 377375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.4836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 11.0382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.2076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.4718,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 5.5191,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-5",
      "model_name": "Kimi-K2.5",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.5",
      "config_revision": "4d01dfe0332d63057c186e0b262165819efb6611",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.5522,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 397375668224,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 19.8688,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 377375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.4836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 11.0382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.2076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.4718,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 5.5191,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-6",
      "model_name": "Kimi-K2.6",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.6",
      "config_revision": "7eb5002f6aadc958aed6a9177b7ed26bb94011bb",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.5522,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 397375668224,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 19.8688,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 377375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.4836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 11.0382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.2076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.4718,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 5.5191,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-7-code",
      "model_name": "Kimi-K2.7-Code",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2.7-Code",
      "config_revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.9422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.4567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 372910511104,
      "utilisation": 0.8632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 18.6455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 352910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.3307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 3.8845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.9422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 10.3586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 336910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 507822961152,
      "utilisation": 0.9918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 507822961152,
      "utilisation": 0.9918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 37.2911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.0717,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.3812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 33.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 93.2276,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 368910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 33.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 93.2276,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 368910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 37.2911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.6448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.6448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.1793,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 300910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 3.8845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct",
      "model_name": "Kimi-K2-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct",
      "config_revision": "fd1984e2b7a3350dbf7305fe73a4ede25c14de50",
      "parameters": 1026408235864,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.9422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.4567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 372910511104,
      "utilisation": 0.8632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 18.6455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 352910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.3307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 3.8845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.9422,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 180910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 10.3586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 336910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 507822961152,
      "utilisation": 0.9918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.8267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 308910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 507822961152,
      "utilisation": 0.9918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 37.2911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.0717,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 192910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.3812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.9134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 33.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 93.2276,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 368910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 33.901,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 93.2276,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 368910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 37.2911,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 62.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 46.6138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 364910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 31.0759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 360910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 23.3069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 356910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 11.6535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 340910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 4.6614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 292910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.6448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 2.6448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 231910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 15.5379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 348910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 7.769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 324910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 5.1793,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 300910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 3.8845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-instruct-0905",
      "model_name": "Kimi-K2-Instruct-0905",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Instruct-0905",
      "config_revision": "ac6c49f04883bd0a0598b790693a72061c676629",
      "parameters": 1026470735448,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 372910511104,
      "utilisation": 1.2948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84910511104
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.5522,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 397375668224,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 19.8688,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 377375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.4836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 237375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.0697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 205375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 11.0382,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 6.209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 501500668224,
      "utilisation": 0.9795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.2076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.4718,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 3.1045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 36.1251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 386375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 99.3439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 393375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 39.7376,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 387375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 66.2293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 391375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 49.672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 389375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 33.1146,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 385375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 24.836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 381375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 12.418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.9672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 2.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 16.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 8.2787,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 5.5191,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 325375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 4.1393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 301375668224
    },
    {
      "model_slug": "moonshotai-kimi-k2-thinking",
      "model_name": "Kimi-K2-Thinking",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K2-Thinking",
      "config_revision": "a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 397375668224,
      "utilisation": 1.3798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109375668224
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 5.7828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 918301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 4.3371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 854301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 3.8552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 822301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 3.8552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 822301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 2.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 678301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 23.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 55.5151,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1090301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 6.9394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 950301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 17.3485,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1046301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 11.5656,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1014301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 5.7828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 918301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 30.8417,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1074301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 2.1686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 598301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 17.3485,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1046301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 17.3485,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1046301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 2.1686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 598301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 111.0301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1100301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 13.8788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1030301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 13.8788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1030301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 6.1683,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 930301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 4.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 840301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 8.6742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 982301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 100.9365,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1099301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 277.5753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1106301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 100.9365,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1099301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 277.5753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1106301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 111.0301,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1100301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 185.0502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1104301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 138.7877,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1102301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 92.5251,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1098301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 69.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 34.6969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 13.8788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1030301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 13.8788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1030301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 7.8745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 969301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 7.8745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 969301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 46.2626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1086301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 23.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 23.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 23.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 23.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 15.4209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1038301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 11.5656,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1014301300736
    },
    {
      "model_slug": "moonshotai-kimi-k3",
      "model_name": "Kimi-K3",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-K3",
      "config_revision": "a590ce090cb049c93a33dfe8c208ec652aa20503",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 1110301300736,
      "utilisation": 3.8552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 822301300736
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.5126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.3844,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.3417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.3417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.2278,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40942331264,
      "utilisation": 0.853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.6151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.8241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.5494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.5126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 35495135232,
      "utilisation": 0.986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.1922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.8241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.8241,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.1922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.876,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.5467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.3645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.7688,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.7054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 4.6899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.7054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 4.6899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.876,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 3.1266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 2.345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.5633,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 18759662336,
      "utilisation": 1.1725,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2759662336
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 30368362496,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.6593,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.698,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.698,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 18759662336,
      "utilisation": 0.7817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40942331264,
      "utilisation": 0.853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40942331264,
      "utilisation": 0.853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40942331264,
      "utilisation": 0.853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 40942331264,
      "utilisation": 0.853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 52740540416,
      "utilisation": 0.5494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "moonshotai-kimi-linear-48b-a3b-instruct",
      "model_name": "Kimi-Linear-48B-A3B-Instruct",
      "publisher": "moonshotai",
      "hf_repo": "moonshotai/Kimi-Linear-48B-A3B-Instruct",
      "config_revision": "e1df551a447157d4658b573f9a695d57658590e9",
      "parameters": 49122681728,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 98411027456,
      "utilisation": 0.3417,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.036,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.4606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0576,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2559,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.9211,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0512,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3782921216,
      "utilisation": 0.9457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.8374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3782921216,
      "utilisation": 0.9457,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.9211,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.9206,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5523757056,
      "utilisation": 0.6905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.7676,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.5757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.2879,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0653,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.0653,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.3838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.1279,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-1-3b",
      "model_name": "Nanbeige4.1-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.1-3B",
      "config_revision": "9c4555c37921c0982af8fffeee1e0a6e6b953c4f",
      "parameters": 3933637120,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9211385856,
      "utilisation": 0.032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4942,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0618,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2746,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0193,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0193,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.9885,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0549,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0772,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 3687604224,
      "utilisation": 0.9219,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8986,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 3687604224,
      "utilisation": 0.9219,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.9885,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5975924736,
      "utilisation": 0.747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.8237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.6178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.3089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1236,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.4119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.2059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.1373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.103,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nanbeige-nanbeige4-2-3b",
      "model_name": "Nanbeige4.2-3B",
      "publisher": "Nanbeige",
      "hf_repo": "Nanbeige/Nanbeige4.2-3B",
      "config_revision": "3384e426066d1a49c3aea90a7190b81260a6533f",
      "parameters": 4169800704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9884983296,
      "utilisation": 0.0343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18592598246,
      "utilisation": 0.9296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.4455,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.8296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.99,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-mini",
      "model_name": "Nex-N2.5-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-mini",
      "config_revision": "87420286149d9cce9bd46cd335ef9bda33c37c1b",
      "parameters": 35107181936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.9379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.9722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 7.2272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.9034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.0151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.8892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.0076,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-5-pro",
      "model_name": "Nex-N2.5-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2.5-Pro",
      "config_revision": "0f389abfca976dcbab5c7638268d01cfe8ace4fe",
      "parameters": 396802360816,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.165,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18592598246,
      "utilisation": 0.9296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.4455,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.3712,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.8296,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.5995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.1392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5569,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.3579,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 3.7343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.4937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 2.4895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.8671,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14937062927,
      "utilisation": 1.2448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2937062927
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14937062927,
      "utilisation": 0.9336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29866392045,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.891,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.5055,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22586040191,
      "utilisation": 0.9411,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38366718471,
      "utilisation": 0.7993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.99,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.7425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-mini",
      "model_name": "Nex-N2-mini",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-mini",
      "config_revision": "ca218dcb1fbe05f84d1807d180cb5d9bcb1c5c93",
      "parameters": 35107181936,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 71279701536,
      "utilisation": 0.2475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.9379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.9722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 7.2272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.9034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.0151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.8892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.0076,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "nex-agi-nex-n2-pro",
      "model_name": "Nex-N2-Pro",
      "publisher": "nex-agi",
      "hf_repo": "nex-agi/Nex-N2-Pro",
      "config_revision": "e7e4a4a29eb46a41a730d0b283ec5af0f16185d9",
      "parameters": 396802360816,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.7845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248094969856,
      "utilisation": 0.9691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 337973039104,
      "utilisation": 0.7823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 7.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.9415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.5691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.7845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.1842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 114632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 436264652800,
      "utilisation": 0.8521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 436264652800,
      "utilisation": 0.8521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 15.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.8368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248094969856,
      "utilisation": 0.9189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 13.6938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 37.6581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 13.6938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 37.6581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 15.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.0683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.0683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.0921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.5691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54632255488
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-405b",
      "model_name": "Hermes-3-Llama-3.1-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-405B",
      "config_revision": "b9fe8ede1f5075f171b2c0a235e20ae4aba33767",
      "parameters": 405853388800,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46004747552,
      "utilisation": 0.9584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46004747552,
      "utilisation": 0.9584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46004747552,
      "utilisation": 0.9584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46004747552,
      "utilisation": 0.9584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46004747552,
      "utilisation": 0.9584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-70b",
      "model_name": "Hermes-3-Llama-3.1-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-70B",
      "config_revision": "1fc86da0ce9cdb14cd775ad270bc7d1b4bf70ede",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8469747840,
      "utilisation": 0.847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.9468,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.9468,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8469747840,
      "utilisation": 0.847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7606728832,
      "utilisation": 0.9508,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.8679,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10414512256,
      "utilisation": 0.6509,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-3-llama-3-1-8b",
      "model_name": "Hermes-3-Llama-3.1-8B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-3-Llama-3.1-8B",
      "config_revision": "896ea440e5a9e6070e3d8a2774daf2b481ab425b",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.7845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248094969856,
      "utilisation": 0.9691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 337973039104,
      "utilisation": 0.7823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 7.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 130632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.9415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.5691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.7845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.1842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 114632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 436264652800,
      "utilisation": 0.8521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.3536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 436264652800,
      "utilisation": 0.8521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 15.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 150632255488,
      "utilisation": 0.8368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248094969856,
      "utilisation": 0.9189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.1768,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 13.6938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 37.6581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 13.6938,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 37.6581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 146632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 15.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 25.1054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 18.829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 12.5527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 9.4145,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 4.7073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.8829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 70632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.0683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.0683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 6.2763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 126632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 3.1382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 2.0921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 150632255488,
      "utilisation": 1.5691,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 54632255488
    },
    {
      "model_slug": "nousresearch-hermes-4-405b",
      "model_name": "Hermes-4-405B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-405B",
      "config_revision": "88e3dce03c4a5535e2f4a2bcc08e939a2b302f82",
      "parameters": 405853388800,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 284350271488,
      "utilisation": 0.9873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-hermes-4-70b",
      "model_name": "Hermes-4-70B",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Hermes-4-70B",
      "config_revision": "d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-llama-2-7b-hf",
      "model_name": "Llama-2-7b-hf",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Llama-2-7b-hf",
      "config_revision": "8efe6c9b93655b934e27bd9981e3ec13e55aee9d",
      "parameters": 6738417664,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nousresearch-meta-llama-3-1-8b-instruct",
      "model_name": "Meta-Llama-3.1-8B-Instruct",
      "publisher": "NousResearch",
      "hf_repo": "NousResearch/Meta-Llama-3.1-8B-Instruct",
      "config_revision": "d10aef7999a2b5ba950ab3974312feeedbfe0b77",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-llama-3-1-nemotron-nano-8b-v1",
      "model_name": "Llama-3.1-Nemotron-Nano-8B-v1",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Llama-3.1-Nemotron-Nano-8B-v1",
      "config_revision": "54641c1611fcff44fa4865626462445e0a153fc7",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4847,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0606,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.9693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0359,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8812,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3729118262,
      "utilisation": 0.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8812,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3729118262,
      "utilisation": 0.9323,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.9693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.9437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5662003985,
      "utilisation": 0.7078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.8078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.6058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0687,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.4039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.2019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.1346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-3-5-content-safety",
      "model_name": "Nemotron-3.5-Content-Safety",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-3.5-Content-Safety",
      "config_revision": "35645ed3543b7e7ffaed2e788699e57a5051497c",
      "parameters": 4300079472,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9693328490,
      "utilisation": 0.0337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19275239428,
      "utilisation": 0.9638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nemotron-cascade-2-30b-a3b",
      "model_name": "Nemotron-Cascade-2-30B-A3B",
      "publisher": "nvidia",
      "hf_repo": "nvidia/Nemotron-Cascade-2-30B-A3B",
      "config_revision": "6327cdbcf907e1c7cec9cb29fb6e6cebdf8feaf7",
      "parameters": 31577937344,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19275239428,
      "utilisation": 0.9638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-5-lightning-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
      "config_revision": "a9904d24bcc1d289a1950fa9d2b978c47cf903b9",
      "parameters": 31577937344,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1483,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19275239428,
      "utilisation": 0.9638,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.3338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.9578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.5388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.356,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.5007,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.2187,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 3.3514,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.3406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 2.2343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.6757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13405690324,
      "utilisation": 1.1171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1405690324
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13405690324,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26834208180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8011,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.4545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23459316126,
      "utilisation": 0.9775,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34480016259,
      "utilisation": 0.7183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.8901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.6675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-30b-a3b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "config_revision": "bf77c3174f68ad409e1c2aa60daeb46e32d1c606",
      "parameters": 31577937344,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 64084332519,
      "utilisation": 0.2225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0467,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.035,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.4485,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0467,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2492,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.8971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0498,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0332,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.8155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3858671358,
      "utilisation": 0.9647,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.8155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3858671358,
      "utilisation": 0.9647,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.8971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.8742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5245442692,
      "utilisation": 0.6557,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.7476,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.5607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0636,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0636,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.3738,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.1246,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-nano-4b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Nano-4B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
      "config_revision": "dfaf35de3e30f1867dd8dbc38a7fc9fb52d3914f",
      "parameters": 3973556832,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8970652222,
      "utilisation": 0.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.69,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.9702,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.8624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.8624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.5749,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.0415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.4996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 62862392049,
      "utilisation": 0.9822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 89346051390,
      "utilisation": 0.9307,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.69,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.3886,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.4851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 62862392049,
      "utilisation": 0.9822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 62862392049,
      "utilisation": 0.9822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.4851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.9991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 76923144674,
      "utilisation": 0.9615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 76923144674,
      "utilisation": 0.9615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 102556978308,
      "utilisation": 0.8012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.5447,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 12.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.5447,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 12.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 45991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.9991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 8.3319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 6.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 4.1659,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 3.1245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.5622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 76923144674,
      "utilisation": 0.9615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 76923144674,
      "utilisation": 0.9615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.9396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 132486294612,
      "utilisation": 0.9396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 2.083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.0415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.0415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.0415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 49991395414,
      "utilisation": 1.0415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1991395414
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 62862392049,
      "utilisation": 0.8731,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 89346051390,
      "utilisation": 0.9307,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-super-120b-a12b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
      "config_revision": "2dc98e2afe4face0e4ce40972a915c45368bd34a",
      "parameters": 123611012096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 248371618452,
      "utilisation": 0.8624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.1629,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 223285205467,
      "utilisation": 0.8722,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 281649827236,
      "utilisation": 0.978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 281649827236,
      "utilisation": 0.978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 401742218248,
      "utilisation": 0.93,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 4.6518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 175285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 11.1643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 203285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.3955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 3.4888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 159285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.3259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.1629,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.2024,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 461648282609,
      "utilisation": 0.9017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 3.4888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 159285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 3.4888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 159285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 461648282609,
      "utilisation": 0.9017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 22.3285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 213285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.7911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 143285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.7911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 143285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.2405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 223285205467,
      "utilisation": 0.827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.7444,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 20.2987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 55.8213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 20.2987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 55.8213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 219285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 22.3285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 213285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 37.2142,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 27.9107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 215285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 18.6071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 211285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 13.9553,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 207285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 6.9777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 191285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.7911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 143285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.7911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 143285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.5836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 1.5836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 9.3036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 199285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 4.6518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 175285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 4.6518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 175285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 4.6518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 175285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 4.6518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 175285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 3.1012,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 151285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 223285205467,
      "utilisation": 2.3259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127285205467
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-3-ultra-550b-a55b-bf16",
      "model_name": "NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
      "config_revision": "77df655d5e9f8362164ed14dd8b48f8bce657498",
      "parameters": 560524578816,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 281649827236,
      "utilisation": 0.978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0737,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0655,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0655,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0437,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.9433,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1965,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.524,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8380540597,
      "utilisation": 0.8381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2358,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2358,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1474,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.9575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4600821926,
      "utilisation": 1.1502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 600821926
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.9575,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4600821926,
      "utilisation": 1.1502,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 600821926
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8380540597,
      "utilisation": 0.8381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5526308596,
      "utilisation": 0.9211,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7430611301,
      "utilisation": 0.9288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.8777,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10532602639,
      "utilisation": 0.6583,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.5895,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2358,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.2358,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1338,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.1965,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "nvidia-nvidia-nemotron-nano-9b-v2",
      "model_name": "NVIDIA-Nemotron-Nano-9B-v2",
      "publisher": "nvidia",
      "hf_repo": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
      "config_revision": "6533e8de2c68e4536bf7c411d7a3ce5734111476",
      "parameters": 8888227328,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18865315759,
      "utilisation": 0.0655,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.6523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.9169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.5433,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43658872384,
      "utilisation": 0.9096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.1829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.7828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59048230144,
      "utilisation": 0.9226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84071406784,
      "utilisation": 0.8757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.6523,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.2127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.4584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59048230144,
      "utilisation": 0.9226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59048230144,
      "utilisation": 0.9226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.4584,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 4.3659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 71938957504,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 71938957504,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.6958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.8693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.9785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 10.9147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 10.9147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 4.3659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 7.2765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 5.4574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 3.6382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 2.7287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.3643,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 71938957504,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 71938957504,
      "utilisation": 0.8992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.8883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 125243716864,
      "utilisation": 0.8883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43658872384,
      "utilisation": 1.8191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19658872384
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43658872384,
      "utilisation": 0.9096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43658872384,
      "utilisation": 0.9096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43658872384,
      "utilisation": 0.9096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43658872384,
      "utilisation": 0.9096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 71938957504,
      "utilisation": 0.9992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84071406784,
      "utilisation": 0.8757,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-120b",
      "model_name": "gpt-oss-120b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-120b",
      "config_revision": "b5c939de8f754692c1647ca79fbf85e8c1e70f8a",
      "parameters": 116829156672,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 234720811264,
      "utilisation": 0.815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 18171785344,
      "utilisation": 0.9086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.4462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.6454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8818,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 2.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 2.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8818,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.4462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-20b",
      "model_name": "gpt-oss-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-20b",
      "config_revision": "6cee5e81ee83917806bbde320786a8fb61efebee",
      "parameters": 20914757184,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0992,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 18171785344,
      "utilisation": 0.9086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.4462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.2231,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.6454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.6693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.0837,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8818,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 2.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 2.2045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 8817922624,
      "utilisation": 0.8818,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.4697,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8817922624,
      "utilisation": 1.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 817922624
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 11482841344,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 15897544384,
      "utilisation": 0.9936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.7261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.3038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 23234854144,
      "utilisation": 0.9681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.8924,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.5949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.4462,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openai-gpt-oss-safeguard-20b",
      "model_name": "gpt-oss-safeguard-20b",
      "publisher": "openai",
      "hf_repo": "openai/gpt-oss-safeguard-20b",
      "config_revision": "8a11e17b25c973a24099d4016bf2e17dd7ec1574",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 42833830144,
      "utilisation": 0.1487,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0696,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0696,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0464,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.5573,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.5564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8851589215,
      "utilisation": 0.8852,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1113,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1565,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8851589215,
      "utilisation": 0.8047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4821408227,
      "utilisation": 1.2054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 821408227
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8851589215,
      "utilisation": 0.8047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4821408227,
      "utilisation": 1.2054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 821408227
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8851589215,
      "utilisation": 0.8852,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5808222090,
      "utilisation": 0.968,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7838713041,
      "utilisation": 0.9798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.9289,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11146257225,
      "utilisation": 0.6966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.626,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2504,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1421,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.1421,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.8346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.4173,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.2087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm-sala",
      "model_name": "MiniCPM-SALA",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM-SALA",
      "config_revision": "9180fe1db74f71fb81bc7105efe88f6b19c959b0",
      "parameters": 9477203968,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20031135945,
      "utilisation": 0.0696,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0124,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.011,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.011,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0073,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.066,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.1584,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0495,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.033,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0165,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.088,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0062,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0495,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0495,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0062,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.3168,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0176,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0117,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0247,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.792,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.792,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.3168,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.528,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.264,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.099,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0396,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.0225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.132,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.066,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.066,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.066,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.066,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.044,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.033,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-1b",
      "model_name": "MiniCPM5-1B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-1B",
      "config_revision": "87179e5c1f455ef22e6223592d2d61351b525bfc",
      "parameters": 1080632832,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3167878528,
      "utilisation": 0.011,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.3096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0967,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0967,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0967,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.6192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0344,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0229,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0484,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.5629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.9581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.5629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.9581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.6192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3832371200,
      "utilisation": 0.6387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.1935,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "openbmb-minicpm5-2b",
      "model_name": "MiniCPM5-2B",
      "publisher": "openbmb",
      "hf_repo": "openbmb/MiniCPM5-2B",
      "config_revision": "abe115e887989b14f05e64a3b260648329324c3f",
      "parameters": 2516756480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 6191667200,
      "utilisation": 0.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.3667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.275,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.2445,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.2445,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.163,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.7904,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.1101,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 2201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.4401,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.5928,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.7335,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.3667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.8208,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.1375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.5928,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.5928,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.1375,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.2201,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 12201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.8801,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.8801,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.3912,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.2608,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.5501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.0183,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 11201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 5.5503,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 18201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.0183,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 11201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 5.5503,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 18201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.2201,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 12201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 3.7002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 16201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 2.7752,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 14201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.8501,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 10201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 22201376640,
      "utilisation": 1.3876,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 6201376640
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29548771168,
      "utilisation": 0.9234,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.8801,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.8801,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.4994,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.4994,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22201376640,
      "utilisation": 0.9251,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.7904,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.7904,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.7904,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37937757760,
      "utilisation": 0.7904,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.9779,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.7335,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-35b",
      "model_name": "Ornith-1.0-35B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-35B",
      "config_revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 70411255680,
      "utilisation": 0.2445,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.9379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.9722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 7.2272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.9034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.7528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.0151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.2585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 419980105728,
      "utilisation": 0.8203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 144544667648,
      "utilisation": 0.803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 240093206528,
      "utilisation": 0.8892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.1293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 13.1404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 36.1362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 14.4545,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 24.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 138544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 18.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 12.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 132544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 9.034,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 4.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.8068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 6.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 3.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 2.0076,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 144544667648,
      "utilisation": 1.5057,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48544667648
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-397b",
      "model_name": "Ornith-1.0-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-397B",
      "config_revision": "5e3e761811e804c295c1d3c0ce68b21da6154209",
      "parameters": 396802360816,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 281091217408,
      "utilisation": 0.976,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0992,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0744,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0661,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0661,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.3967,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.9521,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.119,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.2975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1984,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0992,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.529,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0372,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.2975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.2975,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0372,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8481171904,
      "utilisation": 0.8481,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1488,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.9681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 2751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.9681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 2751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8481171904,
      "utilisation": 0.8481,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6751021536,
      "utilisation": 1.1252,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": 751021536
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7589882304,
      "utilisation": 0.9487,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.8875,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10649413824,
      "utilisation": 0.6656,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.5951,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.238,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1351,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1351,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.7934,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.3967,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.3967,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.3967,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.3967,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.2645,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.1984,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-0-9b",
      "model_name": "Ornith-1.0-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.0-9B",
      "config_revision": "83dc1f5e24ef8527af019a6b3bf66ac0f1c2c999",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19042609344,
      "utilisation": 0.0661,
      "weight_size_evidence": "verified",
      "smallest_format": "Q4_K_M",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.3755,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.2816,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.2504,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.2504,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.1669,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.8091,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19015024210,
      "utilisation": 0.9508,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.4506,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.6068,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.7511,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.3755,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.8401,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.1408,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.6068,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.6068,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.1408,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.9013,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.9013,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.4006,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.267,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5633,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30243350272,
      "utilisation": 0.9451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.9013,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.9013,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5114,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.5114,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22748081920,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.8091,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.8091,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.8091,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.8091,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38836768160,
      "utilisation": 0.5394,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.7511,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-35b-a3b",
      "model_name": "Ornith-1.5-35B-A3B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-35B-A3B",
      "config_revision": "10fbf86fed7ecee4a061f8b499a618f46001cac1",
      "parameters": 35951822704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72101613280,
      "utilisation": 0.2504,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8385,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 245559053088,
      "utilisation": 0.9592,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 287573986080,
      "utilisation": 0.9985,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 287573986080,
      "utilisation": 0.9985,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429770639840,
      "utilisation": 0.9948,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 8.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.0062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8385,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 4.4721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429770639840,
      "utilisation": 0.8394,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429770639840,
      "utilisation": 0.8394,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 16.0995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 245559053088,
      "utilisation": 0.9095,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 14.6359,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 40.2487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 156994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 14.6359,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 40.2487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 156994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 16.0995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.1418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.1418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64994829142
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-397b",
      "model_name": "Ornith-1.5-397B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-397B",
      "config_revision": "8f6cc8a7aea505364523f84ccf37706e8aea0ee7",
      "parameters": 403397928944,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 287573986080,
      "utilisation": 0.9985,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1017,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0763,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0452,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.9765,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1221,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1017,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.5425,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0381,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.3051,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0381,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8680814528,
      "utilisation": 0.8681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1085,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0723,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1526,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.9916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.9916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8680814528,
      "utilisation": 0.8681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7764457408,
      "utilisation": 0.9706,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.909,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10907973216,
      "utilisation": 0.6817,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.6103,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.1385,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.8137,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.4069,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2712,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.2034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ornith-ai-ornith-1-5-9b",
      "model_name": "Ornith-1.5-9B",
      "publisher": "ornith-ai",
      "hf_repo": "ornith-ai/Ornith-1.5-9B",
      "config_revision": "489cb97981b8654bcfcf30ce1f94ed1b62e07b53",
      "parameters": 9653104368,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19529234016,
      "utilisation": 0.0678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.6772,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.9184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.8163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.8163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.5442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43798108160,
      "utilisation": 0.9125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.1899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.8127,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59164669952,
      "utilisation": 0.9244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84281415680,
      "utilisation": 0.8779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.6772,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.2166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.4592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59164669952,
      "utilisation": 0.9244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 59164669952,
      "utilisation": 0.9244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.4592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 4.3798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67230906368,
      "utilisation": 0.8404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67230906368,
      "utilisation": 0.8404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.7224,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.8707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 97273075712,
      "utilisation": 0.7599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.9816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 10.9495,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.9816,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 10.9495,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 4.3798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 7.2997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 5.4748,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 3.6498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 2.7374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.3687,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67230906368,
      "utilisation": 0.8404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67230906368,
      "utilisation": 0.8404,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.9222,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 130028973824,
      "utilisation": 0.9222,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 43798108160,
      "utilisation": 1.8249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19798108160
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43798108160,
      "utilisation": 0.9125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43798108160,
      "utilisation": 0.9125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43798108160,
      "utilisation": 0.9125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 43798108160,
      "utilisation": 0.9125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 67230906368,
      "utilisation": 0.9338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 84281415680,
      "utilisation": 0.8779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-s-2-1",
      "model_name": "Laguna-S-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-S-2.1",
      "config_revision": "0f573140834b11cfac0c2af97a101a7a69a13e22",
      "parameters": 117561977600,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 235102435328,
      "utilisation": 0.8163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.3548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.2661,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.2366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.2366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.1577,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19993716736,
      "utilisation": 0.9997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.4258,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.5707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.7097,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.3548,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.7914,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.1331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.5707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.5707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.1331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.8516,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.8516,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.3785,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.2523,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.5323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.2133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 3.3366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.2133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 3.3366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.3346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 2.2244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.6683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13346375680,
      "utilisation": 1.1122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1346375680
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13346375680,
      "utilisation": 0.8341,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 28491188224,
      "utilisation": 0.8903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.8516,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.8516,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.4832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.4832,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21472758912,
      "utilisation": 0.8947,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 36523436032,
      "utilisation": 0.7609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.7097,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "poolside-laguna-xs-2-1",
      "model_name": "Laguna-XS-2.1",
      "publisher": "poolside",
      "hf_repo": "poolside/Laguna-XS-2.1",
      "config_revision": "c5f36269bbdbd3f27fddc9a9f9dbae0cf2cf57db",
      "parameters": 33442617088,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68128685184,
      "utilisation": 0.2366,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qcri-fanar-1-9b-instruct",
      "model_name": "Fanar-1-9B-Instruct",
      "publisher": "QCRI",
      "hf_repo": "QCRI/Fanar-1-9B-Instruct",
      "config_revision": "d30938d0efb1fd251727ee96df6987802ff84662",
      "parameters": 8783871488,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18098272256,
      "utilisation": 0.9049,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.8086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.3853,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.2428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 3.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.3671,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 2.2785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.7089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13671110656,
      "utilisation": 1.1393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1671110656
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13671110656,
      "utilisation": 0.8544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29110368256,
      "utilisation": 0.9097,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.8669,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.4918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 21842384896,
      "utilisation": 0.9101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37371426816,
      "utilisation": 0.7786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.9632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen-agentworld-35b-a3b",
      "model_name": "Qwen-AgentWorld-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen-AgentWorld-35B-A3B",
      "config_revision": "60d2b0434a53d2e62a7c00a489586815d94ebffb",
      "parameters": 34660610688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69349718016,
      "utilisation": 0.2408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.005,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b",
      "model_name": "Qwen2.5-0.5B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B",
      "config_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
      "parameters": 494032768,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.005,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0135,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.5418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2167,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.3612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.2709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1806,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0452,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-0-5b-instruct",
      "model_name": "Qwen2.5-0.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-0.5B-Instruct",
      "config_revision": "7ae557604adf67be50417f59c2c2f167def9a775",
      "parameters": 494032768,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2167370240,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-1-5b-instruct",
      "model_name": "Qwen2.5-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-1.5B-Instruct",
      "config_revision": "989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
      "parameters": 1543714304,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1248,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.8877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1775,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-14b-instruct",
      "model_name": "Qwen2.5-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-14B-Instruct",
      "config_revision": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8",
      "parameters": 14770033664,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18882531968,
      "utilisation": 0.9441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.7134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.8287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.1338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.5261,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.3805,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.3873,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 3.8151,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.3873,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 3.8151,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.5261,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582432,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582432
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582432,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638752,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798820032,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768369280,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.7134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-32b-instruct",
      "model_name": "Qwen2.5-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-32B-Instruct",
      "config_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
      "parameters": 32763876352,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68481984512,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.2195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-3b-instruct",
      "model_name": "Qwen2.5-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-3B-Instruct",
      "config_revision": "aa8e72537993ba99e69dfaafa59ed015b17504d1",
      "parameters": 3085938688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.3447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.5407,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.8411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.8559,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.0814,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.8272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.8012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 7.7034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.8012,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 7.7034,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.0814,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 5.1356,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 3.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 2.5678,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.9259,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30813613728,
      "utilisation": 0.9629,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.5727,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.5727,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30813613728,
      "utilisation": 1.2839,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6813613728
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.8769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.8411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-72b-instruct",
      "model_name": "Qwen2.5-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-72B-Instruct",
      "config_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
      "parameters": 72706203648,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.086,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0645,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0382,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.3439,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.8254,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.1032,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2579,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.172,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.086,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.4585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2579,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2579,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.9368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0917,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0611,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4285702048,
      "utilisation": 1.0714,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 285702048
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.8517,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4285702048,
      "utilisation": 1.0714,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 285702048
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.9368,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952835680,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523960928,
      "utilisation": 0.9405,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.7807,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368287456,
      "utilisation": 0.5855,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.5159,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2063,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.1171,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.1171,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.6878,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.3439,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.3439,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.3439,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.3439,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.2293,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.172,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-7b-instruct",
      "model_name": "Qwen2.5-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-7B-Instruct",
      "config_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
      "parameters": 7615616512,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507615584,
      "utilisation": 0.0573,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.009,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4178,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2929572864,
      "utilisation": 0.7324,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.4595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-1-5b-instruct",
      "model_name": "Qwen2.5-Coder-1.5B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
      "config_revision": "2e1fd397ee46e1388853d2af2c993145b0f1098a",
      "parameters": 1543714304,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4595457024,
      "utilisation": 0.016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1248,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.9056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.8877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4993,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.0624,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1775,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.1184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2497,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10927339520,
      "utilisation": 0.9934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 2.0047,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9781913600,
      "utilisation": 0.9782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.3365,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2018892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 8018892800,
      "utilisation": 1.0024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18892800
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11397724160,
      "utilisation": 0.9498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14534297600,
      "utilisation": 0.9084,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3995,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.2266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 18111211520,
      "utilisation": 0.7546,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.6658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.4439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.3329,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-14b-instruct",
      "model_name": "Qwen2.5-Coder-14B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-14B-Instruct",
      "config_revision": "aedcc2d42b622764e023cf882b6652e646b95671",
      "parameters": 14770033664,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31957329920,
      "utilisation": 0.111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.3567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.2675,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.1585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 18882531520,
      "utilisation": 0.9441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.7134,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.3567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.8287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.1338,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.1338,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.5261,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.3805,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.2536,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.3873,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 3.8151,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.3873,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 3.8151,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.5261,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 2.5434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.9076,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15260582080,
      "utilisation": 1.2717,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3260582080
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15260582080,
      "utilisation": 0.9538,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29833638080,
      "utilisation": 0.9323,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.4857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.4857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22798819520,
      "utilisation": 0.95,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37768368320,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.9512,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.7134,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-32b-instruct",
      "model_name": "Qwen2.5-Coder-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-32B-Instruct",
      "config_revision": "381fc969f78efac66bc87ff7ddeadb7e73c218a7",
      "parameters": 32763876352,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68483454112,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.2195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-3b-instruct",
      "model_name": "Qwen2.5-Coder-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-3B-Instruct",
      "config_revision": "488639f1ff808d1d3d0ba301aef8c11461451ec5",
      "parameters": 3085938688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0645,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.8254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.4585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.9368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0917,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.8517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4242225152,
      "utilisation": 1.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 242225152
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.8517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4242225152,
      "utilisation": 1.0606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 242225152
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.9368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 5952710656,
      "utilisation": 0.9921,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7523835904,
      "utilisation": 0.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.7807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9368162304,
      "utilisation": 0.5855,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.5159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.1171,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.6878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.3439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.2293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-coder-7b-instruct",
      "model_name": "Qwen2.5-Coder-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-Coder-7B-Instruct",
      "config_revision": "c03e6d358207e414f1eca0bb1891e29f1db0e242",
      "parameters": 7615616512,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16507490304,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.2426,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.2426,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.1618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19678024406,
      "utilisation": 0.9839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.4368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.6019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.7279,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.845,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.1365,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.6019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.6019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.1365,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.6195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.8735,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.8735,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.3882,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.2588,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.546,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.4723,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 4.0487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.4723,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 4.0487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.6195,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.6991,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 2.0243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.3496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16194760109,
      "utilisation": 1.0122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 194760109
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30420528581,
      "utilisation": 0.9506,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.8735,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.8735,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.4956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.4956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23483271117,
      "utilisation": 0.9785,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 38520268009,
      "utilisation": 0.8025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.9706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.7279,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-32b-instruct",
      "model_name": "Qwen2.5-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-32B-Instruct",
      "config_revision": "7cfb30d71a1f4f49a57592323337a4a4727301da",
      "parameters": 33452718336,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 69882191449,
      "utilisation": 0.2426,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0299,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0299,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.4307,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0538,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0897,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2393,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0168,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1346,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0168,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.8615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0479,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0319,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0673,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7831,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3784198676,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7831,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3784198676,
      "utilisation": 0.946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.8615,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.8491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5094562095,
      "utilisation": 0.6368,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.7179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.5384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.2692,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.3589,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1795,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0897,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-3b-instruct",
      "model_name": "Qwen2.5-VL-3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-3B-Instruct",
      "config_revision": "66285546d2b821cf421d4f5eb2576359d3770cd3",
      "parameters": 3754622976,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8614521135,
      "utilisation": 0.0299,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.3447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.9307,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.8411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.7755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.8334,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.9865,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.2908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.8272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.5515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.6308,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.7274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 7.5003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.7274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 7.5003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.0001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 5.0002,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 3.7502,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 2.5001,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.8751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30001321984,
      "utilisation": 0.9375,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.7892,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.5727,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.5727,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30001321984,
      "utilisation": 1.2501,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6001321984
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 47297583104,
      "utilisation": 0.9854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 63136028672,
      "utilisation": 0.8769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 80744044544,
      "utilisation": 0.8411,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-72b-instruct",
      "model_name": "Qwen2.5-VL-72B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-72B-Instruct",
      "config_revision": "89c86200743eec961a297729e7990e8f2ddbc4c5",
      "parameters": 73410777344,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 148904105984,
      "utilisation": 0.517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0413,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.8931,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.4961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8079703914,
      "utilisation": 0.808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.0662,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.917,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4553460044,
      "utilisation": 1.1384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 553460044
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.917,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4553460044,
      "utilisation": 1.1384,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 553460044
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8079703914,
      "utilisation": 0.808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5416881897,
      "utilisation": 0.9028,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7193478603,
      "utilisation": 0.8992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.8406,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10087444766,
      "utilisation": 0.6305,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.5582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2233,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1267,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.7442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.3721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.2481,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.1861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen2-5-vl-7b-instruct",
      "model_name": "Qwen2.5-VL-7B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen2.5-VL-7B-Instruct",
      "config_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5",
      "parameters": 8292166656,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17861351006,
      "utilisation": 0.062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0075,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1625,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.3249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0406,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0406,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0181,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.8123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.8123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.3249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.5415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.4062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.2031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0406,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0406,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.023,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.1354,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0338,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-0-6b",
      "model_name": "Qwen3-0.6B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-0.6B",
      "config_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
      "parameters": 751632384,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3249259520,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.5281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3573950112,
      "utilisation": 0.8935,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.5281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3573950112,
      "utilisation": 0.8935,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b",
      "model_name": "Qwen3-1.7B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B",
      "config_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
      "parameters": 2031739904,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0227,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2905,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0363,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0908,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0323,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0454,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.5281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3904951296,
      "utilisation": 0.9762,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.5281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3904951296,
      "utilisation": 0.9762,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.7262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.4841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.3631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.1815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0726,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.2421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0605,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-1-7b-base",
      "model_name": "Qwen3-1.7B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-1.7B-Base",
      "config_revision": "ea980cb0a6c2ae4b936e82123acc929f1cec04c1",
      "parameters": 1720574976,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5809591296,
      "utilisation": 0.0202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.1238,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.11,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.11,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.0733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.892,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.8802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.0619,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.4951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.0619,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9536943104,
      "utilisation": 0.9537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.176,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.1174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10657705984,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 10657705984,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.9405,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 9536943104,
      "utilisation": 0.9537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 7762192384,
      "utilisation": 1.2937,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1762192384
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 7762192384,
      "utilisation": 0.9703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 11143930240,
      "utilisation": 0.9287,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 14264114528,
      "utilisation": 0.8915,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.9902,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.2247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 17840711008,
      "utilisation": 0.7434,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.6601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.4401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.3301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-14b",
      "model_name": "Qwen3-14B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-14B",
      "config_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
      "parameters": 14768307200,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 31685981184,
      "utilisation": 0.11,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191215296,
      "utilisation": 0.8812,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.9856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.5841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 4.3861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 144531132480,
      "utilisation": 0.9033,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191215296,
      "utilisation": 0.8812,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.4367,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191215296,
      "utilisation": 0.94,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.9345,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.2184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b",
      "model_name": "Qwen3-235B-A22B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B",
      "config_revision": "8efa61729e24bd65b1d152b5ab5409052aa80e65",
      "parameters": 235093634560,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317164096,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.9856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.5841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 4.3861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 144531467264,
      "utilisation": 0.9033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.4367,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.94,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.9345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.2184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-instruct-2507",
      "model_name": "Qwen3-235B-A22B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Instruct-2507",
      "config_revision": "ac9c66cc9b46af7306746a9250f23d47083d689e",
      "parameters": 235093634560,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.9856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.5841,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 4.3861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 144531467264,
      "utilisation": 0.9033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.4367,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670829568,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191549952,
      "utilisation": 0.94,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.9345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544595968,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 21.9305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 7.3102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956166144,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721872384,
      "utilisation": 1.2184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15721872384
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721872384,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-235B-A22B-Thinking-2507",
      "config_revision": "6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5",
      "parameters": 235093634560,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317498368,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910166016,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.9469,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3483,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26697838080,
      "utilisation": 0.8343,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23330887168,
      "utilisation": 0.9721,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089238016,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.8709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b",
      "model_name": "Qwen3-30B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B",
      "config_revision": "ad44e777bcd18fa416d9da3bd8f70d33ebb85d39",
      "parameters": 30532122624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910166016,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3483,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.8709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-instruct-2507",
      "model_name": "Qwen3-30B-A3B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Instruct-2507",
      "config_revision": "0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe",
      "parameters": 30532122624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910166016,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3483,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.8709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-30b-a3b-thinking-2507",
      "model_name": "Qwen3-30B-A3B-Thinking-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-30B-A3B-Thinking-2507",
      "config_revision": "144afc2f379b542fdd4e85a1fcd5e1f79112d95d",
      "parameters": 30532122624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.2675,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.1585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19041522688,
      "utilisation": 0.9521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.7133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.3567,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.8286,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.1337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.1337,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.4975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.3804,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.2536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 3.7438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 3.7438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.4975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975305728,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975305728
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975305728,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29830789760,
      "utilisation": 0.9322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.856,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.4857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709632672,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765202560,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.7133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-32b",
      "model_name": "Qwen3-32B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-32B",
      "config_revision": "9216db5781bf21249d130ec9da846c4624c16137",
      "parameters": 32762123264,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479423488,
      "utilisation": 0.2378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.6288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.6288,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5314220256,
      "utilisation": 0.8857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6288364256,
      "utilisation": 0.786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b",
      "model_name": "Qwen3-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B",
      "config_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
      "parameters": 4022468096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-base",
      "model_name": "Qwen3-4B-Base",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Base",
      "config_revision": "906bfd4b4dc7f14ee4320094d8b41684abff8539",
      "parameters": 4022468096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754595840,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633669120,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6702013440,
      "utilisation": 0.8378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-4b-instruct-2507",
      "model_name": "Qwen3-4B-Instruct-2507",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-4B-Instruct-2507",
      "config_revision": "cdbee75f17c01a7cc42f958dc650907174af0554",
      "parameters": 4022468096,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837539840,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0104,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0556,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0417,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0278,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0741,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0052,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0417,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0417,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0052,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2669,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0148,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2426,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.6672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2426,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.6672,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2669,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.4448,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.3336,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.2224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1668,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0834,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.1112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0556,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0556,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0556,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0556,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0278,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-0-8b",
      "model_name": "Qwen3.5-0.8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-0.8B",
      "config_revision": "2fc06364715b967f1860aea9cf38778875588b17",
      "parameters": 873438784,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 2668948963,
      "utilisation": 0.0093,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.6988,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.9822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.8731,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.8731,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.5347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.8386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 63718283740,
      "utilisation": 0.9956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 90518065724,
      "utilisation": 0.9429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.6988,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.4082,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.4911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 63718283740,
      "utilisation": 0.9956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 63718283740,
      "utilisation": 0.9956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.4911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 5.0694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 77946872775,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 77946872775,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.7454,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.9313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 103886685092,
      "utilisation": 0.8116,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.6085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 12.6734,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.6085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 12.6734,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 5.0694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 8.4489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 6.3367,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 4.2245,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 3.1684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.5842,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 77946872775,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 77946872775,
      "utilisation": 0.9743,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.9516,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 134173253180,
      "utilisation": 0.9516,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 2.1122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 50693652239,
      "utilisation": 1.0561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2693652239
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 63718283740,
      "utilisation": 0.885,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 90518065724,
      "utilisation": 0.9429,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-122b-a10b",
      "model_name": "Qwen3.5-122B-A10B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-122B-A10B",
      "config_revision": "dc4d348443bc740c68e2d77492492c11606384d5",
      "parameters": 125086497008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 251441844125,
      "utilisation": 0.8731,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18550054260,
      "utilisation": 0.9275,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.8622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-27b",
      "model_name": "Qwen3.5-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-27B",
      "config_revision": "fc05daec18b0a78c049392ed2e771dde82bdf654",
      "parameters": 27781427952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0127,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.2736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.057,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.152,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0855,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0107,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.5471,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0304,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0427,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.4974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3339496135,
      "utilisation": 0.8349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.4974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3339496135,
      "utilisation": 0.8349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.5471,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.9119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.6839,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.342,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.0388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.228,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.076,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.057,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-2b",
      "model_name": "Qwen3.5-2B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-2B",
      "config_revision": "15852e8c16360a2fea060d615a32b45270f8a8fc",
      "parameters": 2274069824,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5471436595,
      "utilisation": 0.019,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.3801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1689,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19015024210,
      "utilisation": 0.9508,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.4561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.7601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.3801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.8489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2703,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.7601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-35b-a3b",
      "model_name": "Qwen3.5-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-35B-A3B",
      "config_revision": "59d61f3ce65a6d9863b86d2e96597125219dc754",
      "parameters": 35951822704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8385,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248885152910,
      "utilisation": 0.9722,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282770578942,
      "utilisation": 0.9818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282770578942,
      "utilisation": 0.9818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 430212521971,
      "utilisation": 0.9959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 8.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 140994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.0062,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8385,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 4.4721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 430212521971,
      "utilisation": 0.8403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.5155,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 430212521971,
      "utilisation": 0.8403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 16.0995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 160994829142,
      "utilisation": 0.8944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248885152910,
      "utilisation": 0.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.2578,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 14.6359,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 40.2487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 156994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 14.6359,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 40.2487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 156994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 16.0995,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 26.8325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 154994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 20.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 13.4162,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 148994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 10.0622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 144994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 5.0311,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 128994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.0124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.1418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.1418,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 6.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 136994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 3.3541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 2.236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 160994829142,
      "utilisation": 1.677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64994829142
    },
    {
      "model_slug": "qwen-qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-397B-A17B",
      "config_revision": "8472618112abcbd45acbcdc58436aff4233c23f7",
      "parameters": 403397928944,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282770578942,
      "utilisation": 0.9818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0544,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0242,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.5223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0653,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1088,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0544,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2902,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.6077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0387,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.9496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3982487513,
      "utilisation": 0.9956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.9496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3982487513,
      "utilisation": 0.9956,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.6077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 4948827036,
      "utilisation": 0.8248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6077096870,
      "utilisation": 0.7596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.8705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.6529,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.3264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1306,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0741,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0741,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.4352,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.2176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1451,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.1088,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-4b",
      "model_name": "Qwen3.5-4B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-4B",
      "config_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
      "parameters": 4659865088,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10445720390,
      "utilisation": 0.0363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.0798,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.0473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.4258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.5693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1277,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.3193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2129,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.5677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.0399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.3193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.3193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.0399,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9049524794,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2555,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2555,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.0757,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9049524794,
      "utilisation": 0.8227,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9049524794,
      "utilisation": 0.8227,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4944542162,
      "utilisation": 1.2361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 944542162
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9049524794,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5949671654,
      "utilisation": 0.9916,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 7858573043,
      "utilisation": 0.9823,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11386782689,
      "utilisation": 0.7117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.6386,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2555,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2555,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.1449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.8515,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.4258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.4258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.4258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.4258,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.2129,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-5-9b",
      "model_name": "Qwen3.5-9B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.5-9B",
      "config_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
      "parameters": 9653104368,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 20436568034,
      "utilisation": 0.071,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18550054260,
      "utilisation": 0.9275,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.8622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-27b",
      "model_name": "Qwen3.6-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-27B",
      "config_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
      "parameters": 27781427952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.3801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1689,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19015024210,
      "utilisation": 0.9508,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.4561,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.7601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.3801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.8489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.6135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.1425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.4054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2703,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.3883,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 3.8179,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.5272,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 2.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.9089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 15271540671,
      "utilisation": 1.2726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3271540671
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 15271540671,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 30560053276,
      "utilisation": 0.955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.9121,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.5175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 23104544042,
      "utilisation": 0.9627,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.818,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 39264888348,
      "utilisation": 0.5453,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.7601,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.6-35B-A3B",
      "config_revision": "995ad96eacd98c81ed38be0c5b274b04031597b0",
      "parameters": 35951822704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 72969722133,
      "utilisation": 0.2534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 4.5553,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 682621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 3.4165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 618621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 3.0369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 586621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 3.0369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 586621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 2.0246,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 442621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 18.2213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 826621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 43.7311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 854621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 5.4664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 714621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 13.666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 810621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 9.1106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 778621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 4.5553,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 682621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 24.295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 838621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 1.7082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 13.666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 810621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 13.666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 810621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 1.7082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 362621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 87.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 864621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 10.9328,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 794621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 10.9328,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 794621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 4.859,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 694621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 3.2393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 604621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 746621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 79.5111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 863621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 218.6554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 870621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 79.5111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 863621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 218.6554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 870621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 87.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 864621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 145.7703,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 868621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 109.3277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 866621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 72.8851,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 862621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 54.6639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 858621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 27.3319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 842621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 10.9328,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 794621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 10.9328,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 794621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 733621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 6.203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 733621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 36.4426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 850621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 18.2213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 826621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 18.2213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 826621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 18.2213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 826621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 18.2213,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 826621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 12.1475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 802621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 9.1106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 778621671424
    },
    {
      "model_slug": "qwen-qwen3-8-2-4t-a95b",
      "model_name": "Qwen3.8-2.4T-A95B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-2.4T-A95B",
      "config_revision": "207bd685a7e3696cfaff12ded7c6a7ea0f88c996",
      "parameters": 2446182725504,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 874621671424,
      "utilisation": 3.0369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 586621671424
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18550054260,
      "utilisation": 0.9275,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.8622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-27b",
      "model_name": "Qwen3.8-27B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-27B",
      "config_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
      "parameters": 27781427952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 148945455813,
      "utilisation": 0.7758,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 192527951324,
      "utilisation": 0.7521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 192527951324,
      "utilisation": 0.6685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 192527951324,
      "utilisation": 0.6685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 361277933942,
      "utilisation": 0.8363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.5083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.62,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 148945455813,
      "utilisation": 0.9309,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 91142961767,
      "utilisation": 0.9494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 148945455813,
      "utilisation": 0.7758,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.0111,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 361277933942,
      "utilisation": 0.7056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.1313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 361277933942,
      "utilisation": 0.7056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 7.24,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 72400463698,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 72400463698,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 148945455813,
      "utilisation": 0.8275,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 192527951324,
      "utilisation": 0.7131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 126737958101,
      "utilisation": 0.9901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.5819,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 18.1001,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.5819,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 61400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 18.1001,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 7.24,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 12.0667,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 66400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 9.0501,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 6.0334,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 4.525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 2.2625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 72400463698,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 72400463698,
      "utilisation": 0.905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 129707957795,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 129707957795,
      "utilisation": 0.9199,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 3.0167,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.5083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.5083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.5083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.5083,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 72400463698,
      "utilisation": 1.0056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 400463698
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 91142961767,
      "utilisation": 0.9494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3.8-Flash-Next",
      "config_revision": "de4b8e4d43b917e7706784d8bb445c9af86a3540",
      "parameters": 179999981459,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 192527951324,
      "utilisation": 0.6685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.9198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2874,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8733858592,
      "utilisation": 0.8734,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1437,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5208679424,
      "utilisation": 1.3022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1208679424
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5208679424,
      "utilisation": 1.3022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1208679424
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8733858592,
      "utilisation": 0.8734,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5208679424,
      "utilisation": 0.8681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7859071776,
      "utilisation": 0.9824,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717477664,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.5749,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.23,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.2555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-8b",
      "model_name": "Qwen3-8B",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-8B",
      "config_revision": "b968826d9c46dd6066d109eabc6255188de91218",
      "parameters": 8190735360,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396386304,
      "utilisation": 0.0639,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910166016,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3919,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.9469,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.9797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.1225,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.3483,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816706560,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816706560
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816706560,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698208256,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.7838,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331257344,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089608192,
      "utilisation": 0.7102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.8709,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.6531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-30b-a3b-instruct",
      "model_name": "Qwen3-Coder-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
      "config_revision": "b2cff646eb4bb1d68355c01b18ae02e7cf42d120",
      "parameters": 30532122624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701478912,
      "utilisation": 0.2177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 177009882112,
      "utilisation": 0.9219,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 240023378944,
      "utilisation": 0.9376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 273426703360,
      "utilisation": 0.9494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 273426703360,
      "utilisation": 0.9494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 396960050176,
      "utilisation": 0.9189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 3.6877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 8.8505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 157009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.1063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.8439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 177009882112,
      "utilisation": 0.9219,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 4.9169,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 396960050176,
      "utilisation": 0.7753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.7658,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 396960050176,
      "utilisation": 0.7753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 17.701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 167009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.2126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.2126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 177009882112,
      "utilisation": 0.9834,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 240023378944,
      "utilisation": 0.889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.3829,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 16.0918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 44.2525,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 173009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 16.0918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 166009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 44.2525,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 173009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 17.701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 167009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 29.5016,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 171009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 22.1262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 169009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 14.7508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 165009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 11.0631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 5.5316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.2126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.2126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.2554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.2554,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 7.3754,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 3.6877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 3.6877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 3.6877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 3.6877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 2.4585,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 177009882112,
      "utilisation": 1.8439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81009882112
    },
    {
      "model_slug": "qwen-qwen3-coder-480b-a35b-instruct",
      "model_name": "Qwen3-Coder-480B-A35B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
      "config_revision": "9d90cf8fca1bf7b7acca42d3fc9ae694a2194069",
      "parameters": 480154875392,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 273426703360,
      "utilisation": 0.9494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.8322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.6242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.5548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.5548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.3699,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 46325136384,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.5256,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.9987,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315679136,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.8322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.8476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.3121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315679136,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315679136,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.3121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.0513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133771168,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133771168,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.8877,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.5918,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 7.6281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.7739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 7.6281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.0513,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 5.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 3.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 2.5427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.907,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30512557056,
      "utilisation": 0.9535,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133771168,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133771168,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30512557056,
      "utilisation": 1.2714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6512557056
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 46325136384,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 46325136384,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 46325136384,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 46325136384,
      "utilisation": 0.9651,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133771168,
      "utilisation": 0.9324,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417362336,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-coder-next",
      "model_name": "Qwen3-Coder-Next",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Coder-Next",
      "config_revision": "a7fbcb5c0e12d62a448eaa0e260346bf5dcc0feb",
      "parameters": 79674391296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 159791178752,
      "utilisation": 0.5548,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.8559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.6419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.6905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.5401,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315675488,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.8559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 33809952005,
      "utilisation": 0.9392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315675488,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315675488,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133767520,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133767520,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.9129,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.6086,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.0736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 8.4525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.0736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 8.4525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133767520,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133767520,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133767520,
      "utilisation": 0.9324,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417358688,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-instruct",
      "model_name": "Qwen3-Next-80B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "config_revision": "9c7f2fbe84465e40164a94cc16cd30b6999b0cc7",
      "parameters": 81324862720,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.8559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.6419,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3804,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.6905,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.5401,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315676928,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.8559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 33809952005,
      "utilisation": 0.9392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315676928,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58315676928,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.3209,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133768960,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133768960,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.9129,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.6086,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6751,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.0736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 8.4525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.0736,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 8.4525,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 3.381,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 5.635,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 4.2262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.8175,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 2.1131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.0566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133768960,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133768960,
      "utilisation": 0.8392,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.6129,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33809952005,
      "utilisation": 1.4087,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9809952005
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 42277903336,
      "utilisation": 0.8808,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67133768960,
      "utilisation": 0.9324,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 86417360128,
      "utilisation": 0.9002,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-next-80b-a3b-thinking",
      "model_name": "Qwen3-Next-80B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-Next-80B-A3B-Thinking",
      "config_revision": "e502dd4100cc68c0de57643fd4317ec93a128670",
      "parameters": 81324862720,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 164326191063,
      "utilisation": 0.5706,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191453696,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.9856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.5841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 4.3861,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 144531133600,
      "utilisation": 0.9033,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721776128,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191453696,
      "utilisation": 0.8812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.4367,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670733312,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.3707,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 472670733312,
      "utilisation": 0.9232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 169191453696,
      "utilisation": 0.94,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.9345,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 118544499712,
      "utilisation": 0.9261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 21.9304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.9747,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 21.9304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 8.7722,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 14.6203,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 10.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 7.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 5.4826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 2.7413,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 55721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.0965,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956069888,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 134956069888,
      "utilisation": 0.9571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 3.6551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.8275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 87721776128,
      "utilisation": 1.2184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15721776128
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 87721776128,
      "utilisation": 0.9138,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-235b-a22b-instruct",
      "model_name": "Qwen3-VL-235B-A22B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-235B-A22B-Instruct",
      "config_revision": "710c13861be6c466e66de3f484069440b8f31389",
      "parameters": 235670022896,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 252317165056,
      "utilisation": 0.8761,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0375,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.5996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0222,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0468,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.5451,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3486759777,
      "utilisation": 0.8717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.5451,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3486759777,
      "utilisation": 0.8717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.5996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.9994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.7496,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.3748,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.075,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0425,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.2499,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.1249,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0833,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-2b-instruct",
      "model_name": "Qwen3-VL-2B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-2B-Instruct",
      "config_revision": "89644892e4d85e24eaac8bacfd4f463576704203",
      "parameters": 2127532032,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5996449751,
      "utilisation": 0.0208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3266,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2449,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910116864,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3919,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.6531,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3266,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.9469,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3483,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4447,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4447,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239360,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.8708,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.6531,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-instruct",
      "model_name": "Qwen3-VL-30B-A3B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Instruct",
      "config_revision": "9c4b90e1e4ba969fd3b5378b57d966d725f1b86c",
      "parameters": 31070754032,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3266,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2449,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1451,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 18910116864,
      "utilisation": 0.9455,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3919,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.6531,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3266,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.9469,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.9797,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.1225,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.3483,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2322,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4899,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.1652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 3.2042,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 2.1361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.6021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12816657408,
      "utilisation": 1.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 816657408
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12816657408,
      "utilisation": 0.801,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26698159104,
      "utilisation": 0.8343,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.7838,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4447,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.4447,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23331208192,
      "utilisation": 0.9721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34089239392,
      "utilisation": 0.7102,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.8708,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.6531,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-30b-a3b-thinking",
      "model_name": "Qwen3-VL-30B-A3B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-30B-A3B-Thinking",
      "config_revision": "d0ed0380729be07a546fdefafbb4fe411f341e92",
      "parameters": 31070754032,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 62701110304,
      "utilisation": 0.2177,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.3567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.2675,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.1585,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 19041457152,
      "utilisation": 0.9521,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.428,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.7133,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.3567,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.8286,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.1337,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.5901,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.1337,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.4975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.3804,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.2536,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.535,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 3.7438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.3614,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 3.7438,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.4975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 2.4959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.8719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 14975240192,
      "utilisation": 1.2479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2975240192
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 14975240192,
      "utilisation": 0.936,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 29831089152,
      "utilisation": 0.9322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.4857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.4857,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 22709634080,
      "utilisation": 0.9462,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 37765204000,
      "utilisation": 0.7868,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.9511,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.7133,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-32b-instruct",
      "model_name": "Qwen3-VL-32B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-32B-Instruct",
      "config_revision": "0cfaf48183f594c314753d30a4c4974bc75f3ccb",
      "parameters": 33357390064,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 68479060192,
      "utilisation": 0.2378,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0251,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1693,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0602,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754558976,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3754558976,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.6702,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 5633632256,
      "utilisation": 0.9389,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 6701976576,
      "utilisation": 0.8377,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.9031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.6773,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-4b-instruct",
      "model_name": "Qwen3-VL-4B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-4B-Instruct",
      "config_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17",
      "parameters": 4437815808,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10837502976,
      "utilisation": 0.0376,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0958,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0719,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0426,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.9198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.115,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0958,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.511,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9207959887,
      "utilisation": 0.9208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5479740536,
      "utilisation": 1.3699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1479740536
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5479740536,
      "utilisation": 1.3699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1479740536
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9207959887,
      "utilisation": 0.9208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744352,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717479008,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.2555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.1916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-instruct",
      "model_name": "Qwen3-VL-8B-Instruct",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Instruct",
      "config_revision": "0c351dd01ed87e9c1b53cbc748cba10e6187ff3b",
      "parameters": 8767123696,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004448,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0958,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0719,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0426,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.9198,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.115,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0958,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.511,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.2874,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0359,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9207959887,
      "utilisation": 0.9208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1022,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0681,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1437,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5479740536,
      "utilisation": 1.3699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1479740536
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.9743,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5479740536,
      "utilisation": 1.3699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1479740536
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9207959887,
      "utilisation": 0.9208,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5479740536,
      "utilisation": 0.9133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7035744256,
      "utilisation": 0.8795,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.8931,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10717478912,
      "utilisation": 0.6698,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.5749,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.23,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1305,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.7665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.3833,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.2555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.1916,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "qwen-qwen3-vl-8b-thinking",
      "model_name": "Qwen3-VL-8B-Thinking",
      "publisher": "Qwen",
      "hf_repo": "Qwen/Qwen3-VL-8B-Thinking",
      "config_revision": "92f3c4b4feadd3a016ef468d103bb5f58b2a2c6b",
      "parameters": 8767123696,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18396004352,
      "utilisation": 0.0639,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.0374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.3362,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.8068,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.4482,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.0315,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.0315,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.9454,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.0896,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.0598,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1261,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.8595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4696631613,
      "utilisation": 1.1742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696631613
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.8595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4696631613,
      "utilisation": 1.1742,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 696631613
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.9454,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5438887672,
      "utilisation": 0.9065,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7728030307,
      "utilisation": 0.966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.7878,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9454020686,
      "utilisation": 0.5909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.5043,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2017,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1144,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1144,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.6724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.3362,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.3362,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.3362,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.3362,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.2241,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.1681,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-edge-2603",
      "model_name": "reka-edge-2603",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-edge-2603",
      "config_revision": "492c81c225fbf5f3263a8245b00827721b119a13",
      "parameters": 7128509568,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 16136998406,
      "utilisation": 0.056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.2277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1708,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.9109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.9531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.2733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.6832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.4555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.2277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.6701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.6832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.6832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.0854,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.9637,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.5465,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.5465,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.2429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1619,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3416,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 2.4093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 2.4093,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.9637,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.6062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 9637197824,
      "utilisation": 1.2046,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1637197824
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 9637197824,
      "utilisation": 0.8031,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 14543200256,
      "utilisation": 0.909,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 24125253632,
      "utilisation": 0.7539,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.5465,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.5465,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.3101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 19062339584,
      "utilisation": 0.7943,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.9109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.9109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.9109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.9109,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.6073,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.4555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rekaai-reka-flash-3",
      "model_name": "reka-flash-3",
      "publisher": "RekaAI",
      "hf_repo": "RekaAI/reka-flash-3",
      "config_revision": "8d251ecb02eca9f0f4aa052b29ef922628d2806f",
      "parameters": 20905482240,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 43723630592,
      "utilisation": 0.1518,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "rinna-japanese-gpt-neox-small",
      "model_name": "japanese-gpt-neox-small",
      "publisher": "rinna",
      "hf_repo": "rinna/japanese-gpt-neox-small",
      "config_revision": "84d18c0fa8c9940a61cfc4e25bd9a5686898bac1",
      "parameters": 203611008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-1-70b-euryale-v2-2",
      "model_name": "L3.1-70B-Euryale-v2.2",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.1-70B-Euryale-v2.2",
      "config_revision": "62b47ff8c2c24d326a7b960773b9c853400150b6",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "not_sizable",
      "status_reason": "No published weight file and no usable parameter count for this model.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-3-70b-euryale-v2-3",
      "model_name": "L3.3-70B-Euryale-v2.3",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3.3-70B-Euryale-v2.3",
      "config_revision": "e5737724a37ae00926e95acf663ca73d430dc8ad",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.9466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4978077696,
      "utilisation": 1.2445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 978077696
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8467304448,
      "utilisation": 0.8467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5929013248,
      "utilisation": 0.9882,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7604285440,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.8677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10412068864,
      "utilisation": 0.6508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.7475,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.3738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.2492,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sao10k-l3-8b-lunaris-v1",
      "model_name": "L3-8B-Lunaris-v1",
      "publisher": "Sao10K",
      "hf_repo": "Sao10K/L3-8B-Lunaris-v1",
      "config_revision": "8479c2a7ee119c935b9a02c921cc2a85b698dfe8",
      "parameters": 8030261248,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17940189184,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sapientinc-hrm-text-1b",
      "model_name": "HRM-Text-1B",
      "publisher": "sapientinc",
      "hf_repo": "sapientinc/HRM-Text-1B",
      "config_revision": "9f082d68b8cd0ebc56e33f1c88c45609174c272c",
      "parameters": 1182795264,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.5885,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.8268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.7349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.7349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.4899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 39652313088,
      "utilisation": 0.8261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.9826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.7062,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 60650440704,
      "utilisation": 0.9477,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 87504537600,
      "utilisation": 0.9115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.5885,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.1015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.4134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 60650440704,
      "utilisation": 0.9477,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 60650440704,
      "utilisation": 0.9477,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.4134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 75836442624,
      "utilisation": 0.948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 75836442624,
      "utilisation": 0.948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.6277,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.7839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8827,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.6048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 9.9131,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.6048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 9.9131,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.9652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 6.6087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 4.9565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 31652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 3.3044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 2.4783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.2391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 75836442624,
      "utilisation": 0.948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 75836442624,
      "utilisation": 0.948,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 112991160320,
      "utilisation": 0.8014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 39652313088,
      "utilisation": 1.6522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15652313088
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 39652313088,
      "utilisation": 0.8261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 39652313088,
      "utilisation": 0.8261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 39652313088,
      "utilisation": 0.8261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 39652313088,
      "utilisation": 0.8261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 64854706176,
      "utilisation": 0.9008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 87504537600,
      "utilisation": 0.9115,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-105b",
      "model_name": "sarvam-105b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-105b",
      "config_revision": "fb187764dbad56ed8b26742a802909e858e9bb72",
      "parameters": 106031767424,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 211649054720,
      "utilisation": 0.7349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.3401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.2551,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.2267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.2267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.1511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19365808128,
      "utilisation": 0.9683,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.4081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.5494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.6802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.3401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.9767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.1275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.5494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.5494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.1275,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.3079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.8162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.8162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.3627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.2418,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.5101,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 3.2696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.189,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 3.2696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.3079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 2.1798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.6348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13078505472,
      "utilisation": 1.0899,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1078505472
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13078505472,
      "utilisation": 0.8174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 27375863808,
      "utilisation": 0.8555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.8162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.8162,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.4631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.4631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23830063104,
      "utilisation": 0.9929,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 35160508416,
      "utilisation": 0.7325,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.9069,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.6802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "sarvamai-sarvam-30b",
      "model_name": "sarvam-30b",
      "publisher": "sarvamai",
      "hf_repo": "sarvamai/sarvam-30b",
      "config_revision": "071ae95e933605ca1104a6b4524a36a98488efa4",
      "parameters": 32152650368,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 65294616576,
      "utilisation": 0.2267,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 168319231105,
      "utilisation": 0.8767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 216595155076,
      "utilisation": 0.8461,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 216595155076,
      "utilisation": 0.7521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 216595155076,
      "utilisation": 0.7521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 403517937616,
      "utilisation": 0.9341,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.7402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 4.1766,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 63531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 147010033895,
      "utilisation": 0.9188,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.3052,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 83531056945,
      "utilisation": 0.8701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 168319231105,
      "utilisation": 0.8767,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.3203,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 47531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 403517937616,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.3052,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.3052,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 403517937616,
      "utilisation": 0.7881,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 8.3531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.0441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.0441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 168319231105,
      "utilisation": 0.9351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 216595155076,
      "utilisation": 0.8022,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 7.5937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 20.8828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 7.5937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 20.8828,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 8.3531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 13.9218,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 77531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 10.4414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 6.9609,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 5.2207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 2.6103,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.0441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.0441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.9005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 126971911607,
      "utilisation": 0.9005,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 3.4805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.7402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.7402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.7402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.7402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 83531056945,
      "utilisation": 1.1602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11531056945
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 83531056945,
      "utilisation": 0.8701,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-5-flash",
      "model_name": "Step-3.5-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.5-Flash",
      "config_revision": "ab446a3de5e171ea341227e24bb1f090e1b771f7",
      "parameters": 199384301376,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 216595155076,
      "utilisation": 0.7521,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 169946139496,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 218701716672,
      "utilisation": 0.8543,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 218701716672,
      "utilisation": 0.7594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 218701716672,
      "utilisation": 0.7594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 407481700572,
      "utilisation": 0.9432,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.7566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 4.2158,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 148425221332,
      "utilisation": 0.9277,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.3174,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 84315538799,
      "utilisation": 0.8783,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 169946139496,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.3421,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 407481700572,
      "utilisation": 0.7959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.3174,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.3174,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 407481700572,
      "utilisation": 0.7959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 8.4316,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.0539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.0539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 169946139496,
      "utilisation": 0.9441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 218701716672,
      "utilisation": 0.81,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 121744316941,
      "utilisation": 0.9511,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 21.0789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 21.0789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 8.4316,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 14.0526,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 10.5394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 7.0263,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 5.2697,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 2.6349,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 52315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.0539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.0539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 128188007058,
      "utilisation": 0.9091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 128188007058,
      "utilisation": 0.9091,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 3.5131,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.7566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.7566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.7566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.7566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 84315538799,
      "utilisation": 1.171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12315538799
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 84315538799,
      "utilisation": 0.8783,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "stepfun-ai-step-3-7-flash",
      "model_name": "Step-3.7-Flash",
      "publisher": "stepfun-ai",
      "hf_repo": "stepfun-ai/Step-3.7-Flash",
      "config_revision": "5f6244077ac62e04eec3f320501ff8c2b293373a",
      "parameters": 201365316160,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 218701716672,
      "utilisation": 0.7594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "supralabs-supra2-100m-instruct",
      "model_name": "Supra2-100M-Instruct",
      "publisher": "SupraLabs",
      "hf_repo": "SupraLabs/Supra2-100M-Instruct",
      "config_revision": "2c90ade6ab47b910f48d0157a3e4c421d6311ed5",
      "parameters": 100684032,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.7539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.5654,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.5026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.5026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.3351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46823846368,
      "utilisation": 0.9755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.9047,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61464493264,
      "utilisation": 0.9604,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.7539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.8734,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.2827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61464493264,
      "utilisation": 0.9604,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61464493264,
      "utilisation": 0.9604,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.2827,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.1442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.8041,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.5361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.6137,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.8584,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 7.8605,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.8584,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 7.8605,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.1442,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 5.2403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 3.9302,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 2.6202,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.9651,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 15441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 31441900894,
      "utilisation": 0.9826,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.5572,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.5572,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 31441900894,
      "utilisation": 1.3101,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7441900894
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46823846368,
      "utilisation": 0.9755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46823846368,
      "utilisation": 0.9755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46823846368,
      "utilisation": 0.9755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 46823846368,
      "utilisation": 0.9755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61464493264,
      "utilisation": 0.8537,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78558485451,
      "utilisation": 0.8183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-70b-instruct-2509",
      "model_name": "Apertus-70B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-70B-Instruct-2509",
      "config_revision": "0f2767662a013c0fb631079cea883155252bc3b6",
      "parameters": 70599864480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144745858401,
      "utilisation": 0.5026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0703,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0416,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.3747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.8994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1124,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0937,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.4997,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0351,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8487545801,
      "utilisation": 0.8488,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0666,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1405,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5062863742,
      "utilisation": 1.2657,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062863742
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.9489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5062863742,
      "utilisation": 1.2657,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1062863742
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8487545801,
      "utilisation": 0.8488,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5901417579,
      "utilisation": 0.9836,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7626845283,
      "utilisation": 0.9534,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.8698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10437460307,
      "utilisation": 0.6523,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1276,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.7495,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.3747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.3747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.3747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.3747,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.2498,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.1874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "swiss-ai-apertus-8b-instruct-2509",
      "model_name": "Apertus-8B-Instruct-2509",
      "publisher": "swiss-ai",
      "hf_repo": "swiss-ai/Apertus-8B-Instruct-2509",
      "config_revision": "b946d40447b2b597999b9c86d44bee0b452c919f",
      "parameters": 8053338176,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17987464847,
      "utilisation": 0.0625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.8402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.6301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.5601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.5601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.3734,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 47264904928,
      "utilisation": 0.9847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.5478,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.5458,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58498465528,
      "utilisation": 0.914,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.9096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.8402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.8599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.3151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58498465528,
      "utilisation": 0.914,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 58498465528,
      "utilisation": 0.914,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.3151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.0955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67293913560,
      "utilisation": 0.8412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67293913560,
      "utilisation": 0.8412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.8962,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.5975,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6822,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 7.7388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.8141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 7.7388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.0955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 5.1592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 3.8694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 2.5796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.9347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 30955077304,
      "utilisation": 0.9673,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67293913560,
      "utilisation": 0.8412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67293913560,
      "utilisation": 0.8412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6193,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.6193,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 30955077304,
      "utilisation": 1.2898,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6955077304
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 47264904928,
      "utilisation": 0.9847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 47264904928,
      "utilisation": 0.9847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 47264904928,
      "utilisation": 0.9847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 47264904928,
      "utilisation": 0.9847,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67293913560,
      "utilisation": 0.9346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87321820896,
      "utilisation": 0.9096,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hunyuan-a13b-instruct",
      "model_name": "Hunyuan-A13B-Instruct",
      "publisher": "tencent",
      "hf_repo": "tencent/Hunyuan-A13B-Instruct",
      "config_revision": "290ddb9a56ed23c2c83a1c8081533e58925df952",
      "parameters": 80393183232,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 161316815384,
      "utilisation": 0.5601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0188,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0188,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0339,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0301,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0201,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4927,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3245399104,
      "utilisation": 0.8113,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4927,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3245399104,
      "utilisation": 0.8113,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.5419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.6774,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.4516,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.3387,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0677,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0384,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0384,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.1129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-1-8b",
      "model_name": "Hy-MT2-1.8B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-1.8B",
      "config_revision": "9a341cd1b679d3efd23b46e847b01745a71ed792",
      "parameters": 2038515712,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 5419247056,
      "utilisation": 0.0188,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.3194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.2395,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.2129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.2129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.1419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.6998,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 19842009248,
      "utilisation": 0.9921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.3833,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.9581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.6388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.3194,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.9331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.1198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.9581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.9581,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.1198,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.2538,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.3407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.2271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.1398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 3.1345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.1398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 3.1345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.2538,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 2.0897,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.5672,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12537952256,
      "utilisation": 1.0448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 537952256
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 12537952256,
      "utilisation": 0.7836,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26130741248,
      "utilisation": 0.8166,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.4349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22821967872,
      "utilisation": 0.9509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.6998,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.6998,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.6998,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 33591035744,
      "utilisation": 0.6998,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.8517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.6388,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-30b-a3b",
      "model_name": "Hy-MT2-30B-A3B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-30B-A3B",
      "config_revision": "d3ead4dba61c09aac60a261a96ad1df3e705febb",
      "parameters": 30064725888,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 61321308160,
      "utilisation": 0.2129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.3737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.8969,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0934,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.4983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2803,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.9856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0664,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1401,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.896,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4977655480,
      "utilisation": 1.2444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 977655480
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.896,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4977655480,
      "utilisation": 1.2444,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 977655480
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.9856,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5928554008,
      "utilisation": 0.9881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7603732216,
      "utilisation": 0.9505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.8213,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 9855670720,
      "utilisation": 0.616,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.5606,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2242,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.7474,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.3737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.3737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.3737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.3737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.2491,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.1869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy-mt2-7b",
      "model_name": "Hy-MT2-7B",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy-MT2-7B",
      "config_revision": "9b0eb4e8f001def3e5ff6469a0ac96fdb39ec223",
      "parameters": 8029540352,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 17938727448,
      "utilisation": 0.0623,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 181153466368,
      "utilisation": 0.9435,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 5.5007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 148681000960,
      "utilisation": 0.9293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 181153466368,
      "utilisation": 0.9435,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.0559,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 11.0014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 100013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 168950499328,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.9055,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 10.0013,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 27.5035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 10.0013,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 27.5035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 11.0014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 100013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.7802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.7802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14013810688
    },
    {
      "model_slug": "tencent-hy3",
      "model_name": "Hy3",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3",
      "config_revision": "a960ebc3da325ba167f069f76c41eb62c9280d22",
      "parameters": 298786155776,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 181153466368,
      "utilisation": 0.9435,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.955,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 5.5007,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 148681000960,
      "utilisation": 0.9293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 181153466368,
      "utilisation": 0.9435,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.0559,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 74013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.719,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 46013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 315562723328,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 11.0014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 100013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 168950499328,
      "utilisation": 0.9386,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.9055,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.8595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 10.0013,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 27.5035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 10.0013,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 27.5035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 106013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 11.0014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 100013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 18.3356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 13.7517,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 102013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 9.1678,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 98013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 6.8759,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 94013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 3.4379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.3752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.7802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 110013810688,
      "utilisation": 0.7802,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 4.5839,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 2.292,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 62013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.528,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 110013810688,
      "utilisation": 1.146,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14013810688
    },
    {
      "model_slug": "tencent-hy3-preview",
      "model_name": "Hy3-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy3-preview",
      "config_revision": "549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a",
      "parameters": 298786155776,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 244485621760,
      "utilisation": 0.8489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.4596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.0947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 280240384000,
      "utilisation": 0.9731,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 280240384000,
      "utilisation": 0.9731,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 381435555840,
      "utilisation": 0.883,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 5.8383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 14.012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.7515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 4.3788,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 216240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.9192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 184240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.4596,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 7.7845,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 244240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 466162554880,
      "utilisation": 0.9105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 4.3788,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 216240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 4.3788,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 216240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 466162554880,
      "utilisation": 0.9105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 28.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 270240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 3.503,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 3.503,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.5569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 100240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.0379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.1894,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 152240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 25.4764,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 70.0601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 25.4764,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 269240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 70.0601,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 276240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 28.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 270240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 46.7067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 274240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 35.03,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 272240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 23.3534,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 268240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 17.515,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 8.7575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 248240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 3.503,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 3.503,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 200240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.9875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 1.9875,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 11.6767,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 5.8383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 5.8383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 5.8383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 5.8383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 232240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 3.8922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 208240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 280240384000,
      "utilisation": 2.9192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 184240384000
    },
    {
      "model_slug": "tencent-hy4-preview",
      "model_name": "Hy4-preview",
      "publisher": "tencent",
      "hf_repo": "tencent/Hy4-preview",
      "config_revision": "705d81ee51566a186d645b74c974d642ef2828fe",
      "parameters": 779960992733,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 280240384000,
      "utilisation": 0.9731,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.3999,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.2999,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.2666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.2666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.1777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.8786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 16704852672,
      "utilisation": 0.8352,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.4798,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.659,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.7997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.3999,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 33235653312,
      "utilisation": 0.9232,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.15,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.659,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.659,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.15,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.9597,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.9597,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.4265,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.2844,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5998,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.5186,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 4.1762,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.5186,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 4.1762,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.6705,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.7841,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 2.0881,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.3921,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 16704852672,
      "utilisation": 1.0441,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 704852672
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 29158003392,
      "utilisation": 0.9112,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.9597,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.9597,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.5445,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 23890450432,
      "utilisation": 0.9954,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.8786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.8786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.8786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.8786,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 42174763712,
      "utilisation": 0.5858,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.7997,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-skyfall-36b-v2",
      "model_name": "Skyfall-36B-v2",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/Skyfall-36B-v2",
      "config_revision": "8d5f36e2fbf323fc17157ba8d8bc9716927825a5",
      "parameters": 36910535680,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 76775380992,
      "utilisation": 0.2666,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.1388,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.1041,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.0925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.0925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.0617,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.5551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.7582,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.1665,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.4164,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.1388,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.7402,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.4164,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.4164,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.052,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9619380672,
      "utilisation": 0.9619,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.3331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.3331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.148,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.0987,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2082,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9882,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.7333,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9882,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.7333,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 9619380672,
      "utilisation": 0.9619,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6933223872,
      "utilisation": 1.1555,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 933223872
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 6933223872,
      "utilisation": 0.8667,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 10869807552,
      "utilisation": 0.9058,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.9478,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.8327,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.3331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.3331,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.189,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 15164545472,
      "utilisation": 0.6319,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.5551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.5551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.5551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.5551,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.3701,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.2776,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thedrummer-unslopnemo-12b-v4-1",
      "model_name": "UnslopNemo-12B-v4.1",
      "publisher": "TheDrummer",
      "hf_repo": "TheDrummer/UnslopNemo-12B-v4.1",
      "config_revision": "c20fadc8281ea8237acb8054f4a84524ca257478",
      "parameters": 12247782400,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 26646452672,
      "utilisation": 0.0925,
      "weight_size_evidence": "verified",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.3815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 353661390336,
      "utilisation": 0.8187,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 7.3679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 17.6831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 333661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.2104,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 193661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 5.526,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 289661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 3.684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 257661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 161661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 9.8239,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 317661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 481647106560,
      "utilisation": 0.9407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 5.526,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 289661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 5.526,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 289661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 481647106560,
      "utilisation": 0.9407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 35.3661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 343661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 4.4208,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 273661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 4.4208,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 273661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.9648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 173661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.3099,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 225661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 32.151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 342661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 88.4153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 32.151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 342661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 88.4153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 349661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 35.3661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 343661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 58.9436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 347661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 44.2077,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 29.4718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 22.1038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 337661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 11.0519,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 321661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 4.4208,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 273661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 4.4208,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 273661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.5082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 2.5082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 212661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 14.7359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 7.3679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 7.3679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 7.3679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 7.3679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 4.912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 3.684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 257661390336
    },
    {
      "model_slug": "thinkingmachines-inkling",
      "model_name": "Inkling",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling",
      "config_revision": "828496eeae4c243ff1a22f7f28ff83694f2f7bc9",
      "parameters": 952377623626,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 353661390336,
      "utilisation": 1.228,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65661390336
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 187813636074,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 220626011747,
      "utilisation": 0.8618,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.9897,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.9897,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.6598,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.2402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 5.3764,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 87528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 156962689139,
      "utilisation": 0.981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.6801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.1201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 187813636074,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.9869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 71528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.5567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.6801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.6801,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.5567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 10.7528,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.3441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.3441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 165473295190,
      "utilisation": 0.9193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 220626011747,
      "utilisation": 0.8171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 107528036024,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 9.7753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 26.882,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 9.7753,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 26.882,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 10.7528,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 17.9213,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 101528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 13.441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 8.9607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 6.7205,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 3.3603,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 75528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.3441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.3441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 135220750244,
      "utilisation": 0.959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 135220750244,
      "utilisation": 0.959,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 4.4803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.2402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.2402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.2402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 2.2402,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.4934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 107528036024,
      "utilisation": 1.1201,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 11528036024
    },
    {
      "model_slug": "thinkingmachines-inkling-small",
      "model_name": "Inkling-Small",
      "publisher": "thinkingmachines",
      "hf_repo": "thinkingmachines/Inkling-Small",
      "config_revision": "8cc5877b44d343f88b92086aa1fb72897950f06a",
      "parameters": 265956439090,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 285020714561,
      "utilisation": 0.9897,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0481,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0428,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0428,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.2566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.6159,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.077,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1925,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0642,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3422,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0241,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1925,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1925,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0241,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.8431,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.154,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.154,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0684,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0962,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7664,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5663287562,
      "utilisation": 1.4158,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1663287562
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7664,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5663287562,
      "utilisation": 1.4158,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1663287562
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.8431,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5663287562,
      "utilisation": 0.9439,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 7426638330,
      "utilisation": 0.9283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8430639046,
      "utilisation": 0.7026,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.7699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.3849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.154,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.154,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0874,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.5133,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.2566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.2566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.2566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.2566,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1711,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.1283,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiger-lab-vlm2vec-full",
      "model_name": "VLM2Vec-Full",
      "publisher": "TIGER-Lab",
      "hf_repo": "TIGER-Lab/VLM2Vec-Full",
      "config_revision": "a0dcd8153e2563f1b39d80ae9afd71a4394773fb",
      "parameters": 4146621440,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 12318096646,
      "utilisation": 0.0428,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0598,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.7654,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0957,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.2392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4252,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0299,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.2392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.2392,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0299,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.8542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1913,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1913,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.085,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1196,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3725116009,
      "utilisation": 0.9313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7765,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 3725116009,
      "utilisation": 0.9313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.8542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 5903805168,
      "utilisation": 0.984,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 6794225954,
      "utilisation": 0.8493,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 8541688025,
      "utilisation": 0.7118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.9567,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.4784,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1913,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1913,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1086,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1086,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.6378,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.3189,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.2126,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.1595,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tiiuae-falcon-7b",
      "model_name": "falcon-7b",
      "publisher": "tiiuae",
      "hf_repo": "tiiuae/falcon-7b",
      "config_revision": "ec89142b67d748a1865ea4451372db8313ada0d8",
      "parameters": 7217189760,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 15307803425,
      "utilisation": 0.0532,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "tinyllama-tinyllama-1-1b-chat-v1-0",
      "model_name": "TinyLlama-1.1B-Chat-v1.0",
      "publisher": "TinyLlama",
      "hf_repo": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
      "config_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
      "parameters": 1100048384,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 2048 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0051,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.003,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0652,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0081,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0362,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0025,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1185,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1185,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.326,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.2173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.1087,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0092,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0092,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0181,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "titanml-tiny-mixtral",
      "model_name": "tiny-mixtral",
      "publisher": "TitanML",
      "hf_repo": "TitanML/tiny-mixtral",
      "config_revision": "40b934a7efa1bf6732ba19ed7a97956c201fd775",
      "parameters": 246961152,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 1303896064,
      "utilisation": 0.0045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0221,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.4769,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0596,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.149,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.2649,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.149,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.149,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.9537,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0353,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.867,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3703718409,
      "utilisation": 0.9259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.867,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3703718409,
      "utilisation": 0.9259,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.9537,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.9324,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5594203616,
      "utilisation": 0.6993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.7948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.5961,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.298,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.3974,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.1325,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-4b",
      "model_name": "NeoHorse-1-4B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-4B",
      "config_revision": "56f0584bb40578a2c33b1b40a08ccd17243ad710",
      "parameters": 4205751296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9537095456,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0744,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0661,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0661,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0441,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.3966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.9519,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.119,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.2975,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5288,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.2975,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.2975,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0372,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8475223763,
      "utilisation": 0.8475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.238,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.238,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1058,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1487,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.9676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4667618925,
      "utilisation": 1.1669,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 667618925
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.9676,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4667618925,
      "utilisation": 1.1669,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 667618925
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8475223763,
      "utilisation": 0.8475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5599933689,
      "utilisation": 0.9333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7518286039,
      "utilisation": 0.9398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.8869,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10643163378,
      "utilisation": 0.6652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.5949,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.238,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.238,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.7932,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.3966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.3966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.3966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.3966,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.2644,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.1983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "tokenrhythm-neohorse-1-9b",
      "model_name": "NeoHorse-1-9B",
      "publisher": "TokenRhythm",
      "hf_repo": "TokenRhythm/NeoHorse-1-9B",
      "config_revision": "ba5b6e40d88a6ddf4591e176738254a3bc715765",
      "parameters": 8953803264,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19037353938,
      "utilisation": 0.0661,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1321,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18550054260,
      "utilisation": 0.9275,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2973,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.8622,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.8919,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1115,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.3171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.2114,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.446,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.1361,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 3.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.2497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 2.0829,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.5621,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12497175645,
      "utilisation": 1.0414,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 497175645
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15389916830,
      "utilisation": 0.9619,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.9699,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7135,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.4048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21342087769,
      "utilisation": 0.8893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31037806124,
      "utilisation": 0.6466,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.7928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.5946,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "ukisai-swift-1-5-qwen3-8-27b",
      "model_name": "Swift-1.5-Qwen3.8-27b",
      "publisher": "ukisai",
      "hf_repo": "ukisai/Swift-1.5-Qwen3.8-27b",
      "config_revision": "b4c84d42903a8646b25857eb2827288d92ed85a4",
      "parameters": 27781427952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 57082894829,
      "utilisation": 0.1982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "undi95-remm-slerp-l2-13b",
      "model_name": "ReMM-SLERP-L2-13B",
      "publisher": "Undi95",
      "hf_repo": "Undi95/ReMM-SLERP-L2-13B",
      "config_revision": "9cf491450795628e28b8fd73f411977f7908c623",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.3082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.2311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.2054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.2054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.137,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.6779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 19655184384,
      "utilisation": 0.9828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.3698,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.9245,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.3082,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.9038,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.1156,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.9245,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.9245,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.1156,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.3095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.7396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.7396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.3287,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.2191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4622,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.1904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 3.2737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.1904,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 3.2737,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.3095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 2.1825,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.6368,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13094719488,
      "utilisation": 1.0912,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1094719488
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13094719488,
      "utilisation": 0.8184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 25657454592,
      "utilisation": 0.8018,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.7396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.7396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.4196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 22565376000,
      "utilisation": 0.9402,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.6779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.6779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.6779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 32537014272,
      "utilisation": 0.6779,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.8218,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.6163,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-27b-it",
      "model_name": "gemma-2-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-27b-it",
      "config_revision": "c3d7bab17f14fbda4ad59f1dec764eb7562c7e55",
      "parameters": 27227128320,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 59167567872,
      "utilisation": 0.2054,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0299,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0177,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.3828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0478,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2126,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.015,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.7655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0425,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0284,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3875417088,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3875417088,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.7655,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4651367424,
      "utilisation": 0.7752,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.9569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.6379,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.4784,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.2392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0543,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.1063,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0797,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-2b-it",
      "model_name": "gemma-2-2b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-2b-it",
      "config_revision": "457f2e15bf550c227ce6ad86e2ec108d3e42c106",
      "parameters": 2614341888,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7655046144,
      "utilisation": 0.0266,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.088,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.0783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.0783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.0522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.4695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.6507,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1409,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.3522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1174,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.6261,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.3522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.3522,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.044,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9490677760,
      "utilisation": 0.9491,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1252,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.0835,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1761,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.9595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.5641,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.9595,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.5641,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9490677760,
      "utilisation": 0.9491,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6256361472,
      "utilisation": 1.0427,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 256361472
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7426121728,
      "utilisation": 0.9283,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 10554236928,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 13014523904,
      "utilisation": 0.8134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.7043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2817,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.1598,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.9391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.4695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.4695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.4695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.4695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.313,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.2348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-2-9b-it",
      "model_name": "gemma-2-9b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-2-9b-it",
      "config_revision": "fc7d4737cda11c3a19af2b722319e846670b4d89",
      "parameters": 9241705984,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 22538215424,
      "utilisation": 0.0783,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.0888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.0888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.0592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.5331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.7081,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1599,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.2665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1333,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.05,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3998,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.05,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9908973509,
      "utilisation": 0.9909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1422,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.0948,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9908973509,
      "utilisation": 0.9008,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.5072,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2028833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9908973509,
      "utilisation": 0.9008,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.5072,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2028833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 9908973509,
      "utilisation": 0.9909,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 6028833900,
      "utilisation": 1.0048,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28833900
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 7297839120,
      "utilisation": 0.9122,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 11211493873,
      "utilisation": 0.9343,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.8851,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.7996,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.1815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 14162349948,
      "utilisation": 0.5901,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.5331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.5331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.5331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.5331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.3554,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.2665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-12b-it",
      "model_name": "gemma-3-12b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-12b-it",
      "config_revision": "9478e665381f42974aa06177b019352fb6291876",
      "parameters": 12187325040,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 25587967173,
      "utilisation": 0.0888,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0134,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0214,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0357,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0952,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0536,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0067,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.3428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.019,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0127,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0268,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.3428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.5714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.4285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2857,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.2143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0429,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0243,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.1428,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0476,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0357,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-1b-it",
      "model_name": "gemma-3-1b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-1b-it",
      "config_revision": "5b11413a10db4e486ef16a20101fd028f8f2499c",
      "parameters": 999885952,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 3428106752,
      "utilisation": 0.0119,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.2272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.1346,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.6568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18661156992,
      "utilisation": 0.9331,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.3636,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.9089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.6059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.303,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.8758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.1136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.9089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.9089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.1136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.2149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.7271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.7271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.3232,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.2154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.1045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 3.0373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.1045,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 3.0373,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.2149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 2.0249,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.5186,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12149149056,
      "utilisation": 1.0124,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149149056
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 15612274176,
      "utilisation": 0.9758,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.9852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.7271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.7271,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.4125,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 21562347648,
      "utilisation": 0.8984,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.6568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.6568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.6568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 31527255552,
      "utilisation": 0.6568,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.8079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.6059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-27b-it",
      "model_name": "gemma-3-27b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-27b-it",
      "config_revision": "7a5a3053dbd5d1d58e48159e87b9df2fc545a49a",
      "parameters": 27432406640,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 58168776192,
      "utilisation": 0.202,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0221,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.4773,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0597,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1492,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2652,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1492,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1492,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.9547,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.053,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0746,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.8679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 3943524298,
      "utilisation": 0.9859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.8679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 3943524298,
      "utilisation": 0.9859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.9547,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.9192,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5515203345,
      "utilisation": 0.6894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.7955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.5967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.2983,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1193,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0677,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.3978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1989,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.1326,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0994,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-gemma-3-4b-it",
      "model_name": "gemma-3-4b-it",
      "publisher": "unsloth",
      "hf_repo": "unsloth/gemma-3-4b-it",
      "config_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
      "parameters": 4300079472,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 9546527850,
      "utilisation": 0.0331,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0159,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0848,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0254,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0636,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0424,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1131,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0636,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0636,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0226,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0318,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2666164224,
      "utilisation": 0.6665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3701,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 2666164224,
      "utilisation": 0.6665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.4071,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.6785,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.5089,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.3392,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.2544,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0289,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0289,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.1696,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0848,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0848,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0848,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0848,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0424,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-1b-instruct",
      "model_name": "Llama-3.2-1B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-1B-Instruct",
      "config_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
      "parameters": 1235814400,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4070928384,
      "utilisation": 0.0141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.035,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0207,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1866,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.4479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.14,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0933,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0467,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.2489,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.14,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.14,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0175,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.8959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0498,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0332,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.07,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.8144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3978082304,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.8144,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3978082304,
      "utilisation": 0.9945,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.8959,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.9296,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5577605120,
      "utilisation": 0.6972,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.7466,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.5599,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.28,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.112,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.3733,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1866,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1866,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1866,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1866,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.1244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0933,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-2-3b-instruct",
      "model_name": "Llama-3.2-3B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.2-3B-Instruct",
      "config_revision": "006f5dcd1393c3add266de40994ba96225e9689d",
      "parameters": 3212749824,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8958771200,
      "utilisation": 0.0311,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-3-3-70b-instruct",
      "model_name": "Llama-3.3-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-3.3-70B-Instruct",
      "config_revision": "99cd0d2c829e92a67c844f9144c2509632e5c87f",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 161437790161,
      "utilisation": 0.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248932856528,
      "utilisation": 0.9724,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282665894164,
      "utilisation": 0.9815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282665894164,
      "utilisation": 0.9815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429444766257,
      "utilisation": 0.9941,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 3.3633,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 8.0719,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.5225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.6816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 161437790161,
      "utilisation": 0.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 4.4844,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429444766257,
      "utilisation": 0.8388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.5225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.5225,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 429444766257,
      "utilisation": 0.8388,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 16.1438,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 151437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 161437790161,
      "utilisation": 0.8969,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 248932856528,
      "utilisation": 0.922,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.2612,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 14.6762,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 40.3594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 157437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 14.6762,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 150437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 40.3594,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 157437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 16.1438,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 151437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 26.9063,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 155437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 20.1797,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 153437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 13.4531,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 149437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 10.0899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 145437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 5.0449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.1449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.1449,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 6.7266,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 3.3633,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 3.3633,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 3.3633,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 3.3633,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 2.2422,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 161437790161,
      "utilisation": 1.6816,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65437790161
    },
    {
      "model_slug": "unsloth-llama-4-maverick-17b-128e-instruct",
      "model_name": "Llama-4-Maverick-17B-128E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Maverick-17B-128E-Instruct",
      "config_revision": "86c5ecb6f2fbddf604c60120125fb202e09f2556",
      "parameters": 401583781376,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 282665894164,
      "utilisation": 0.9815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.6143,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.8586,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.7632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.7632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.5088,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 45432762976,
      "utilisation": 0.9465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.2716,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.7371,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 56745089728,
      "utilisation": 0.8866,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 91632685677,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.6143,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.262,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.4293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 56745089728,
      "utilisation": 0.8866,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 56745089728,
      "utilisation": 0.8866,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.4293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 4.5433,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 78229004400,
      "utilisation": 0.9779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 78229004400,
      "utilisation": 0.9779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.6552,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.814,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.9214,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 4.1303,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 11.3582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 4.1303,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 11.3582,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 41432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 4.5433,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 7.5721,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 39432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 5.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 37432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 3.7861,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 33432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 2.8395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.4198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 78229004400,
      "utilisation": 0.9779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 78229004400,
      "utilisation": 0.9779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.8364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 117937579937,
      "utilisation": 0.8364,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 45432762976,
      "utilisation": 1.893,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21432762976
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 45432762976,
      "utilisation": 0.9465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 45432762976,
      "utilisation": 0.9465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 45432762976,
      "utilisation": 0.9465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 45432762976,
      "utilisation": 0.9465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 69103093743,
      "utilisation": 0.9598,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 91632685677,
      "utilisation": 0.9545,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-llama-4-scout-17b-16e-instruct",
      "model_name": "Llama-4-Scout-17B-16E-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Llama-4-Scout-17B-16E-Instruct",
      "config_revision": "afd8e498c87bda51c7ea8ec68ea2f7c066e6340b",
      "parameters": 108641793536,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 219789261377,
      "utilisation": 0.7632,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5648,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.3347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.4569,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.9037,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.7531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.8094,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.9589,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.2824,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.8033,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5356,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 18138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 7.2847,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.9139,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 4.8565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 3.6423,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 2.4282,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.8212,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 29138718720,
      "utilisation": 0.9106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.9807,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.5564,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 29138718720,
      "utilisation": 1.2141,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5138718720
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 45960335360,
      "utilisation": 0.9575,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 61370028032,
      "utilisation": 0.8524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 78456934400,
      "utilisation": 0.8173,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "unsloth-meta-llama-3-1-70b-instruct",
      "model_name": "Meta-Llama-3.1-70B-Instruct",
      "publisher": "unsloth",
      "hf_repo": "unsloth/Meta-Llama-3.1-70B-Instruct",
      "config_revision": "1fdd0a465a29664d155aee8a9f77c55a65cc8d5f",
      "parameters": 70553706496,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 144599797760,
      "utilisation": 0.5021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 177833199616,
      "utilisation": 0.9262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 205146769408,
      "utilisation": 0.8014,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 264925038592,
      "utilisation": 0.9199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 264925038592,
      "utilisation": 0.9199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 264925038592,
      "utilisation": 0.6133,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 4.6129,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 72257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 152126310400,
      "utilisation": 0.9508,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.4415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 92257601536,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 177833199616,
      "utilisation": 0.9262,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.5627,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 496324790272,
      "utilisation": 0.9694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.4415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.4415,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 496324790272,
      "utilisation": 0.9694,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 9.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.1532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.1532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 177833199616,
      "utilisation": 0.988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 264925038592,
      "utilisation": 0.9812,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.9751,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 8.3871,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 23.0644,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 8.3871,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 23.0644,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 88257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 9.2258,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 82257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 15.3763,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 86257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 11.5322,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 84257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 7.6881,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 5.7661,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 76257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 2.8831,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 60257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.1532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.1532,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 12257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.8852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 124812740608,
      "utilisation": 0.8852,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 3.8441,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 68257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.922,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 44257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 92257601536,
      "utilisation": 1.2814,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20257601536
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 92257601536,
      "utilisation": 0.961,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "upstage-solar-open2-250b",
      "model_name": "Solar-Open2-250B",
      "publisher": "upstage",
      "hf_repo": "upstage/Solar-Open2-250B",
      "config_revision": "9190fbe63a2ad8e17fc766ccceb36de7c66f004b",
      "parameters": 250287810304,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 264925038592,
      "utilisation": 0.9199,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0187,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.4043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0505,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0421,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2246,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1263,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0299,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0632,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.735,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 3795869696,
      "utilisation": 0.949,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.8085,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.826,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4956188672,
      "utilisation": 0.6195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.6738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.5053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.2527,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1011,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0573,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.3369,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1684,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.1123,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0842,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "webai-official-twil-lm3",
      "model_name": "TwIL-LM3",
      "publisher": "webAI-Official",
      "hf_repo": "webAI-Official/TwIL-LM3",
      "config_revision": "7857a949111bebd3849ee22a2a10b6225a14081d",
      "parameters": 3075098624,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 8085204992,
      "utilisation": 0.0281,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0309,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3951,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0494,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0412,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.2195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1235,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0154,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0439,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0617,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7184,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 3895415808,
      "utilisation": 0.9739,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.7903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 4718093312,
      "utilisation": 0.7863,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.9878,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.6586,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.4939,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.247,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0988,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.056,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.3293,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1646,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.1098,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0823,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "weiboai-vibethinker-3b",
      "model_name": "VibeThinker-3B",
      "publisher": "WeiboAI",
      "hf_repo": "WeiboAI/VibeThinker-3B",
      "config_revision": "77bd2cced09193c8b9a59a32bd8577bbd1f3e01c",
      "parameters": 3085938688,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 7902651392,
      "utilisation": 0.0274,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0113,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.2448,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.051,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0255,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.136,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0765,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0096,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.4896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0272,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0181,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.4451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3044419584,
      "utilisation": 0.7611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.4451,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 3044419584,
      "utilisation": 0.7611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.4896,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.8161,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.6121,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.408,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.306,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.153,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0612,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.0347,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.102,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.068,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.051,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-1-7b",
      "model_name": "Spark-X2.5-1.7B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-1.7B",
      "config_revision": "448e61eb392c00f2c403185c5b56d5e0665bfaab",
      "parameters": 1707657216,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 4896466944,
      "utilisation": 0.017,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0393,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0233,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.5028,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0629,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0524,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2794,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1571,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0196,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.5889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0559,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0372,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0786,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.9143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3886043136,
      "utilisation": 0.9715,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.9143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 3886043136,
      "utilisation": 0.9715,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.5889,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.9815,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 5888700416,
      "utilisation": 0.7361,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.6285,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.3143,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1257,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0713,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0713,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.2095,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1397,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.1048,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xhtoken-spark-x2-5-4b",
      "model_name": "Spark-X2.5-4B",
      "publisher": "XHToken",
      "hf_repo": "XHToken/Spark-X2.5-4B",
      "config_revision": "5e10fcc0286756aebf7c41dc52c1e42d95c70281",
      "parameters": 4112079360,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 10056790016,
      "utilisation": 0.0349,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 225151412224,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.7791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 5.7789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 157248415744,
      "utilisation": 0.9828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.2105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.6053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5",
      "model_name": "MiMo-V2.5",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5",
      "config_revision": "63651580ca774f8504f676040460aed3e1244ac1",
      "parameters": 310775040000,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.4755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.8744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 18.8869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.3609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 10.4927,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.0985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 197737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.2464,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-5-pro",
      "model_name": "MiMo-V2.5-Pro",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.5-Pro",
      "config_revision": "21d1ecfecd7bd70f31be25ca49d7edd21f003659",
      "parameters": 1023244718976,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.9975,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1247,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.5542,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.3117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.885,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0739,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4848199075,
      "utilisation": 1.212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 848199075
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.8045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 4848199075,
      "utilisation": 1.212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 848199075
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8849722369,
      "utilisation": 0.885,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 5827995931,
      "utilisation": 0.9713,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7844048525,
      "utilisation": 0.9805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.9273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 11128073522,
      "utilisation": 0.6955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.6234,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2494,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.1415,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.8312,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.4156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2771,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.2078,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-distill-qwen-9b",
      "model_name": "MiMo-V2.6-Distill-Qwen-9B",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B",
      "config_revision": "2367e865d009c13ac81713a2878291d33ab28177",
      "parameters": 9409813744,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 19949773907,
      "utilisation": 0.0693,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 225151412224,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.7791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 5.7789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 157248415744,
      "utilisation": 0.9828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.2105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.6053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-mopd",
      "model_name": "MiMo-V2.6-Flash-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-MOPD",
      "config_revision": "2479e2d0029eca9a34cc7e7f55a121925f81908e",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 225151412224,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.7791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 5.7789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 157248415744,
      "utilisation": 0.9828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.2105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.6053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-flash-rl",
      "model_name": "MiMo-V2.6-Flash-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
      "config_revision": "5711b268169967567844e1e560e8a3966da959b1",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.4755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.8744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 18.8869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.3609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 10.4927,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.0985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 197737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.2464,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-mopd",
      "model_name": "MiMo-V2.6-Pro-MOPD",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-MOPD",
      "config_revision": "adea8e2c5373181e5a973fa1ecb343cb31af214b",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.4755,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.8744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 18.8869,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 357737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.3609,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 217737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.9674,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 185737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 10.4927,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 341737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.9022,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 313737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 377737906176,
      "utilisation": 0.7378,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.0985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 197737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.399,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.9511,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 249737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 34.3398,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 366737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 94.4345,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 373737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 37.7738,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 367737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 62.9563,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 371737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 47.2172,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 369737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 31.4782,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 365737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 23.6086,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 361737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 11.8043,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 345737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 4.7217,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 297737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 2.679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 236737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 15.7391,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 353737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 7.8695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 329737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 5.2464,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 305737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 3.9348,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 281737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-6-pro-rl",
      "model_name": "MiMo-V2.6-Pro-RL",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
      "config_revision": "73875d00b30a89ef8cc353a0b60b0e9f9561952d",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 377737906176,
      "utilisation": 1.3116,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 89737906176
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 225151412224,
      "utilisation": 0.8795,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.7791,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 5.7789,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 157248415744,
      "utilisation": 0.9828,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.9319,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.2105,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 79578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.8059,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 336571295744,
      "utilisation": 0.6574,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 178921728000,
      "utilisation": 0.994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9635,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.903,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 10.5072,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 28.8947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 11.5579,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 105578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 19.2631,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 14.4473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 9.6316,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 103578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 7.2237,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 3.6118,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.4447,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 115578759168,
      "utilisation": 0.8197,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 4.8158,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 91578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 2.4079,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.6053,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 43578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 115578759168,
      "utilisation": 1.2039,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 19578759168
    },
    {
      "model_slug": "xiaomimimo-mimo-v2-flash",
      "model_name": "MiMo-V2-Flash",
      "publisher": "XiaomiMiMo",
      "hf_repo": "XiaomiMiMo/MiMo-V2-Flash",
      "config_revision": "1afd314a2406c282e0956375c34a676501c78649",
      "parameters": 309785318400,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 260131743744,
      "utilisation": 0.9032,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.3314,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.2486,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.221,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.221,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.1473,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.7161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19340733574,
      "utilisation": 0.967,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.3977,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.9943,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.6629,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.3314,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.9547,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.9943,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.9943,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.1243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.3539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.7954,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.7954,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.3535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.2357,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4971,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.2308,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 3.3847,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.2308,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 3.3847,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.3539,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 2.2564,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.6923,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13538639671,
      "utilisation": 1.1282,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1538639671
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13538639671,
      "utilisation": 0.8462,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26812831641,
      "utilisation": 0.8379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.7954,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.7954,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4513,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.4513,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23476725193,
      "utilisation": 0.9782,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.7161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.7161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.7161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34370771043,
      "utilisation": 0.7161,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.8838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.6629,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "xingchen-agi-xing4-0-29b-a4b",
      "model_name": "Xing4.0-29B-A4B",
      "publisher": "XingChen-AGI",
      "hf_repo": "XingChen-AGI/Xing4.0-29B-A4B",
      "config_revision": "baae3c3e813cad5f888f1f485cfff659c89076c5",
      "parameters": 31215031088,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63634862688,
      "utilisation": 0.221,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.8527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.6396,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.5685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.5685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.379,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 41737279460,
      "utilisation": 0.8695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.6637,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 13273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.547,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 59152897818,
      "utilisation": 0.9243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.9117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.8527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 33273329582,
      "utilisation": 0.9243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 59152897818,
      "utilisation": 0.9243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 59152897818,
      "utilisation": 0.9243,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.3198,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 3.3273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67840385388,
      "utilisation": 0.848,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67840385388,
      "utilisation": 0.848,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.9096,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.6064,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6838,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 3.0248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 8.3183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 3.0248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 8.3183,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 3.3273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 23273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 5.5456,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 27273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 4.1592,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 25273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.7728,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 21273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 2.0796,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.0398,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67840385388,
      "utilisation": 0.848,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67840385388,
      "utilisation": 0.848,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.6207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 33273329582,
      "utilisation": 1.3864,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9273329582
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 41737279460,
      "utilisation": 0.8695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 41737279460,
      "utilisation": 0.8695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 41737279460,
      "utilisation": 0.8695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 41737279460,
      "utilisation": 0.8695,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 67840385388,
      "utilisation": 0.9422,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 87521863077,
      "utilisation": 0.9117,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "yandex-aliceai-foundation-80b-a3b-base",
      "model_name": "AliceAI-Foundation-80B-A3B-Base",
      "publisher": "yandex",
      "hf_repo": "yandex/AliceAI-Foundation-80B-A3B-Base",
      "config_revision": "84105ba3dc09f9e8141e69b74761af9a4d848f48",
      "parameters": 81286433408,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 163727894397,
      "utilisation": 0.5685,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.2892,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.2169,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1285,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.6293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 18070355176,
      "utilisation": 0.9035,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.347,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.8675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.5783,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.2892,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.8391,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.8675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.8675,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.2188,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.3084,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.2056,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.4337,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 3.0469,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.108,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 3.0469,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.2188,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 2.0313,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 6187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.5235,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 12187730176,
      "utilisation": 1.0156,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 187730176
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 14999105176,
      "utilisation": 0.9374,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.944,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.3938,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 23669480176,
      "utilisation": 0.9862,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.6293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.6293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.6293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 30206855176,
      "utilisation": 0.6293,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.7711,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.5783,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "youssofal-qwen3-8-27b-mtplx-optimized-speed",
      "model_name": "Qwen3.8-27B-MTPLX-Optimized-Speed",
      "publisher": "Youssofal",
      "hf_repo": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed",
      "config_revision": "123db8bcc7101455b00d9aad36c0e760c6e7de02",
      "parameters": null,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 55519355176,
      "utilisation": 0.1928,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183100695616,
      "utilisation": 0.9536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 253961993845,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.8911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 7.2894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 145788773097,
      "utilisation": 0.9112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.5186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183100695616,
      "utilisation": 0.9536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.7518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.7518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 14.5789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 145788773097,
      "utilisation": 0.8099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 13.2535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 36.4472,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 13.2535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 36.4472,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 14.5789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.034,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.034,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.0248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.5186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49788773097
    },
    {
      "model_slug": "zai-org-glm-4-5",
      "model_name": "GLM-4.5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5",
      "config_revision": "cbb2c7cfb52fa128a9660cb1a7a78e017899e115",
      "parameters": 358337791296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.8726,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.7756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.7756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.5171,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 46089158505,
      "utilisation": 0.9602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.3045,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 26089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.7488,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 57591724891,
      "utilisation": 0.8999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 93066026265,
      "utilisation": 0.9694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.624,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.2803,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 10089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.4363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 57591724891,
      "utilisation": 0.8999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 57591724891,
      "utilisation": 0.8999,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.4363,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 4.6089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 79436935002,
      "utilisation": 0.993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 79436935002,
      "utilisation": 0.993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.6656,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.8273,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.936,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 4.1899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 11.5223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 4.1899,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 11.5223,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 42089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 4.6089,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 7.6815,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 40089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 5.7611,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 38089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 3.8408,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 2.8806,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.4403,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 79436935002,
      "utilisation": 0.993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 79436935002,
      "utilisation": 0.993,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 119813290478,
      "utilisation": 0.8497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 46089158505,
      "utilisation": 1.9204,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 22089158505
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 46089158505,
      "utilisation": 0.9602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 46089158505,
      "utilisation": 0.9602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 46089158505,
      "utilisation": 0.9602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 46089158505,
      "utilisation": 0.9602,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 70157553716,
      "utilisation": 0.9744,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 93066026265,
      "utilisation": 0.9694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5-air",
      "model_name": "GLM-4.5-Air",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5-Air",
      "config_revision": "a24ceef6ce4f3536971efe9b778bdaa1bab18daa",
      "parameters": 110468824832,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 223377813758,
      "utilisation": 0.7756,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.5996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.4966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.0473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.7195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 89439080448,
      "utilisation": 0.9317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.5996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.1374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 4.0947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.6396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 10.2366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 10.2366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 4.0947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 66290343936,
      "utilisation": 0.9207,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 89439080448,
      "utilisation": 0.9317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-5v",
      "model_name": "GLM-4.5V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.5V",
      "config_revision": "ed47433b37111465ec527affaaddceff371bca04",
      "parameters": 107710933120,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 176934967296,
      "utilisation": 0.9215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 252593412096,
      "utilisation": 0.9867,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 252593412096,
      "utilisation": 0.8771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 252593412096,
      "utilisation": 0.8771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 376730802176,
      "utilisation": 0.8721,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 6.5533,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 111065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 131065993216,
      "utilisation": 0.8192,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.3653,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 176934967296,
      "utilisation": 0.9215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 3.6407,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 95065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 376730802176,
      "utilisation": 0.7358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.0479,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 67065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 376730802176,
      "utilisation": 0.7358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 13.1066,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 176934967296,
      "utilisation": 0.983,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 252593412096,
      "utilisation": 0.9355,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.024,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 11.9151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 32.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 11.9151,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 32.7665,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 127065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 13.1066,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 21.8443,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 16.3832,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 123065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 10.9222,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 119065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 8.1916,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 115065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 4.0958,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 99065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.6383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 51065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 131065993216,
      "utilisation": 0.9295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 131065993216,
      "utilisation": 0.9295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 5.4611,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 107065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 2.7305,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 83065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.8204,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 59065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 131065993216,
      "utilisation": 1.3653,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 35065993216
    },
    {
      "model_slug": "zai-org-glm-4-6",
      "model_name": "GLM-4.6",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6",
      "config_revision": "be72194883d968d7923a07e2f61681ea9a2826d1",
      "parameters": 356785898816,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 252593412096,
      "utilisation": 0.8771,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.5996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.8381,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.4966,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.0473,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 20946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.7195,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 89439080448,
      "utilisation": 0.9317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.5996,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.1374,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 62258350080,
      "utilisation": 0.9728,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.419,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 4.0947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.6396,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7946,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8994,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 10.2366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.7224,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 29946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 10.2366,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 36946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 4.0947,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 30946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 6.8244,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 34946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 5.1183,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 3.4122,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 28946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 2.5592,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 24946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.2796,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 8946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 77513973760,
      "utilisation": 0.9689,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 115122833408,
      "utilisation": 0.8165,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 40946536448,
      "utilisation": 1.7061,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 16946536448
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 40946536448,
      "utilisation": 0.8531,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 66290343936,
      "utilisation": 0.9207,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 89439080448,
      "utilisation": 0.9317,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v",
      "model_name": "GLM-4.6V",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V",
      "config_revision": "4e2d47eb0b41c5280d8294b17cef9e94fdcfff46",
      "parameters": 107710933120,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 214543812608,
      "utilisation": 0.7449,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1132,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0503,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.4527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.604,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1358,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.3395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1132,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6036,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0424,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.3395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.3395,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0424,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.9588,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2716,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2716,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1207,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0805,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1698,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.8717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5211484199,
      "utilisation": 1.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1211484199
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.8717,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5211484199,
      "utilisation": 1.3029,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1211484199
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.9588,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5211484199,
      "utilisation": 0.8686,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_K_M",
      "required_bytes": 7454023091,
      "utilisation": 0.9318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 9588487819,
      "utilisation": 0.799,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 12080626564,
      "utilisation": 0.755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.6791,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2716,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2716,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.1541,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.9054,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.4527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.4527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.4527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.4527,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.3018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.2264,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-6v-flash",
      "model_name": "GLM-4.6V-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.6V-Flash",
      "config_revision": "411bb4d77144a3f03accbf4b780f5acb8b7cde4e",
      "parameters": 10292777472,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 21730105444,
      "utilisation": 0.0755,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183100695616,
      "utilisation": 0.9536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 253961993845,
      "utilisation": 0.992,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.8911,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 7.2894,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 125788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 145788773097,
      "utilisation": 0.9112,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.5186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 183100695616,
      "utilisation": 0.9536,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.0497,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 109788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.7518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.2779,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 81788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 384934456563,
      "utilisation": 0.7518,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 14.5789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 145788773097,
      "utilisation": 0.8099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9625,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.139,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 17788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 13.2535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 36.4472,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 13.2535,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 134788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 36.4472,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 141788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 14.5789,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 135788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 24.2981,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 139788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 18.2236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 137788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 12.1491,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 133788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 9.1118,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 4.5559,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 113788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.8224,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 65788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.034,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.034,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 4788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 6.0745,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 121788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 3.0373,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 97788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 2.0248,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 73788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 145788773097,
      "utilisation": 1.5186,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 49788773097
    },
    {
      "model_slug": "zai-org-glm-4-7",
      "model_name": "GLM-4.7",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7",
      "config_revision": "602d01efcdd332c5238ca4bcede555defbe83eb7",
      "parameters": 358337791296,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 259874567401,
      "utilisation": 0.9023,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.3318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.2489,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.2212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.2212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.1475,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.7176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 19410551313,
      "utilisation": 0.9705,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.3982,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.9955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.6637,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.3318,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.9568,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.9955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.9955,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.1244,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.3607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.7964,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.7964,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.354,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.236,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4978,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 3.4018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.237,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 2607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 3.4018,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 9607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.3607,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 3607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 2.2679,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 7607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.7009,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 5607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 13607257124,
      "utilisation": 1.1339,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1607257124
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 13607257124,
      "utilisation": 0.8505,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 26884195141,
      "utilisation": 0.8401,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.7964,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.7964,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4519,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.4519,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 23547398549,
      "utilisation": 0.9811,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.7176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.7176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.7176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 34443698063,
      "utilisation": 0.7176,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.8849,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.6637,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-4-7-flash",
      "model_name": "GLM-4.7-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-4.7-Flash",
      "config_revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67",
      "parameters": 31221488576,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 63713843603,
      "utilisation": 0.2212,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 418807304192,
      "utilisation": 0.9695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 13.5107,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 250214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.6888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 7.506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 234214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.5012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 198214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5",
      "model_name": "GLM-5",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5",
      "config_revision": "c183ef8c61faee82855eca1ed9bb3a9a7ce3b0b2",
      "parameters": 753864139008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 418807304192,
      "utilisation": 0.9695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 13.5107,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 250214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.6888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 7.506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 234214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.5012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 198214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-1",
      "model_name": "GLM-5.1",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.1",
      "config_revision": "26e1bd6e011feb778d25ae34b09b07074139d92d",
      "parameters": 753864139008,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 418807304192,
      "utilisation": 0.9695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 13.5107,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 250214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.6888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 7.506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 234214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.5012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 198214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-2",
      "model_name": "GLM-5.2",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.2",
      "config_revision": "cf457fa734ab149ffef225f80893eb38c6ff5cdc",
      "parameters": 753329940480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0555,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 14214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 418807304192,
      "utilisation": 0.9695,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 13.5107,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 250214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.6888,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 110214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.4074,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 78214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 7.506,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 234214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 4.2221,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 206214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_0",
      "required_bytes": 511217094656,
      "utilisation": 0.9985,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.5012,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 90214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.0008,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.1111,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 142214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 24.565,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 259214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 67.5537,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 266214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 27.0215,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 260214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 45.0358,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 264214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 33.7769,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 262214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 22.5179,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 258214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 16.8884,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 254214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 8.4442,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 238214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.3777,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 190214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 1.9164,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 129214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 11.259,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 246214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 5.6295,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 222214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 3.753,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 198214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 270214922240,
      "utilisation": 2.8147,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 174214922240
    },
    {
      "model_slug": "zai-org-glm-5-3",
      "model_name": "GLM-5.3",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3",
      "config_revision": "187fb9fff6319062325ff825627ef6db084d9bc6",
      "parameters": 753329940480,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 270214922240,
      "utilisation": 0.9382,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 188041157930,
      "utilisation": 0.9794,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 230616459589,
      "utilisation": 0.9008,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 264957858569,
      "utilisation": 0.92,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 264957858569,
      "utilisation": 0.92,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 342758197544,
      "utilisation": 0.7934,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.6732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 6.4158,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 108315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 128315239470,
      "utilisation": 0.802,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.0049,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.3366,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q4_0",
      "required_bytes": 188041157930,
      "utilisation": 0.9794,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 3.5643,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 92315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 342758197544,
      "utilisation": 0.6694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.0049,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.0049,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 64315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 342758197544,
      "utilisation": 0.6694,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 12.8315,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.6039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.6039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q3_K_M",
      "required_bytes": 161773000114,
      "utilisation": 0.8987,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 264957858569,
      "utilisation": 0.9813,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.0025,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 11.665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 32.0788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 11.665,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 117315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 32.0788,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 124315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 12.8315,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 118315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 21.3859,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 122315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 16.0394,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 120315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 10.6929,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 116315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 8.0197,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 112315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 4.0099,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 96315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.6039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.6039,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 48315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 128315239470,
      "utilisation": 0.91,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 128315239470,
      "utilisation": 0.91,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 5.3465,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 104315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.6732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.6732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.6732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 2.6732,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 80315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.7822,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 56315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 128315239470,
      "utilisation": 1.3366,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 32315239470
    },
    {
      "model_slug": "zai-org-glm-5-3-flash",
      "model_name": "GLM-5.3-Flash",
      "publisher": "zai-org",
      "hf_repo": "zai-org/GLM-5.3-Flash",
      "config_revision": "03eb5366286afd40d2221b1d9c63a6dd1ba4832e",
      "parameters": 321323031390,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 264957858569,
      "utilisation": 0.92,
      "weight_size_evidence": "bounded",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0718,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0426,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.9191,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1149,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0957,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5106,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2872,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0359,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8649013248,
      "utilisation": 0.8649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1021,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0681,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1436,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.9679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5105596416,
      "utilisation": 1.2764,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1105596416
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.9679,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": false,
      "best_format": null,
      "required_bytes": 5105596416,
      "utilisation": 1.2764,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": 1105596416
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q6_K",
      "required_bytes": 8649013248,
      "utilisation": 0.8649,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q2_K",
      "required_bytes": 5105596416,
      "utilisation": 0.8509,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q5_K_M",
      "required_bytes": 7771371520,
      "utilisation": 0.9714,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.8873,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "Q8_0",
      "required_bytes": 10647107584,
      "utilisation": 0.6654,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.5744,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2298,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1304,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.7659,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.383,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.2553,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.1915,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zed-industries-zeta-2-1",
      "model_name": "zeta-2.1",
      "publisher": "zed-industries",
      "hf_repo": "zed-industries/zeta-2.1",
      "config_revision": "9a83b321df711c0d207a1c23a92534337798c2ab",
      "parameters": 8250462208,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "ok",
      "status_reason": null,
      "fits": true,
      "best_format": "FP16",
      "required_bytes": 18381666304,
      "utilisation": 0.0638,
      "weight_size_evidence": "derived",
      "smallest_format": "Q2_K",
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-instinct-mi300x",
      "device_name": "AMD Instinct MI300X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 5300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-instinct-mi325x",
      "device_name": "AMD Instinct MI325X Accelerator",
      "vendor": "AMD",
      "device_memory_gb": 256,
      "device_bandwidth_gb_s": 6000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-instinct-mi350x",
      "device_name": "AMD Instinct MI350X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-instinct-mi355x",
      "device_name": "AMD Instinct MI355X",
      "vendor": "AMD",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 8000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-instinct-mi455x",
      "device_name": "AMD Instinct MI455X",
      "vendor": "AMD",
      "device_memory_gb": 432,
      "device_bandwidth_gb_s": 23300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-ai-pro-r9700",
      "device_name": "AMD Radeon™ AI PRO R9700",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-pro-w7800",
      "device_name": "AMD Radeon PRO W7800",
      "vendor": "AMD",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-pro-w7900",
      "device_name": "AMD Radeon PRO W7900",
      "vendor": "AMD",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6600",
      "device_name": "AMD Radeon RX 6600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6600-xt",
      "device_name": "AMD Radeon RX 6600 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6650-xt",
      "device_name": "AMD Radeon RX 6650 XT",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 280,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6700-xt",
      "device_name": "AMD Radeon RX 6700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6750-xt",
      "device_name": "AMD Radeon RX 6750 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6800",
      "device_name": "AMD Radeon RX 6800",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6800-xt",
      "device_name": "AMD Radeon RX 6800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6900-xt",
      "device_name": "AMD Radeon RX 6900 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-6950-xt",
      "device_name": "AMD Radeon RX 6950 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7600",
      "device_name": "AMD Radeon RX 7600",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7600-xt",
      "device_name": "AMD Radeon RX 7600 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7700-xt",
      "device_name": "AMD Radeon RX 7700 XT",
      "vendor": "AMD",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7800-xt",
      "device_name": "AMD Radeon RX 7800 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 624,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7900-xt",
      "device_name": "AMD Radeon RX 7900 XT",
      "vendor": "AMD",
      "device_memory_gb": 20,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-7900-xtx",
      "device_name": "AMD Radeon™ RX 7900 XTX",
      "vendor": "AMD",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-9060-xt-16gb",
      "device_name": "AMD Radeon RX 9060 XT 16GB",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-9060-xt-8gb",
      "device_name": "AMD Radeon RX 9060 XT 8GB",
      "vendor": "AMD",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-9070",
      "device_name": "AMD Radeon RX 9070",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-radeon-rx-9070-xt",
      "device_name": "AMD Radeon™ RX 9070 XT",
      "vendor": "AMD",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 640,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-ryzen-ai-max-plus-395-strix-halo",
      "device_name": "AMD Ryzen AI Max+ 395 with Radeon 8060S",
      "vendor": "AMD",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "amd-ryzen-ai-max-plus-pro-495",
      "device_name": "AMD Ryzen AI Max+ PRO 495 with Radeon 8065S",
      "vendor": "AMD",
      "device_memory_gb": 160,
      "device_bandwidth_gb_s": 273.06,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m1-max",
      "device_name": "Apple M1 Max",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m1-pro",
      "device_name": "Apple M1 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m1-ultra",
      "device_name": "Apple M1 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m2",
      "device_name": "Apple M2",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m2-max",
      "device_name": "Apple M2 Max",
      "vendor": "Apple",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m2-pro",
      "device_name": "Apple M2 Pro",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m2-ultra",
      "device_name": "Apple M2 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 192,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m3",
      "device_name": "Apple M3",
      "vendor": "Apple",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 100,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m3-max",
      "device_name": "Apple M3 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 400,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m3-pro",
      "device_name": "Apple M3 Pro",
      "vendor": "Apple",
      "device_memory_gb": 36,
      "device_bandwidth_gb_s": 150,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m3-ultra",
      "device_name": "Apple M3 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m4",
      "device_name": "Apple M4",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 120,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m4-max",
      "device_name": "Apple M4 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 546,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m4-pro",
      "device_name": "Apple M4 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m5",
      "device_name": "Apple M5",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 153,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m5-max",
      "device_name": "Apple M5 Max",
      "vendor": "Apple",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 614,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m5-pro",
      "device_name": "Apple M5 Pro",
      "vendor": "Apple",
      "device_memory_gb": 64,
      "device_bandwidth_gb_s": 307,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m5-ultra",
      "device_name": "Apple M5 Ultra",
      "vendor": "Apple",
      "device_memory_gb": 512,
      "device_bandwidth_gb_s": 1200,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "apple-m6",
      "device_name": "Apple M6",
      "vendor": "Apple",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 170,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-a750",
      "device_name": "Intel Arc A750",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-a770-16gb",
      "device_name": "Intel Arc A770 (16GB)",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 560,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-a770-8gb",
      "device_name": "Intel Arc A770 (8GB)",
      "vendor": "Intel",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-b570",
      "device_name": "Intel Arc B570",
      "vendor": "Intel",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 380,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-b580",
      "device_name": "Intel Arc B580",
      "vendor": "Intel",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-pro-b50-16gb",
      "device_name": "Intel Arc Pro B50 16GB",
      "vendor": "Intel",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-pro-b60-24gb",
      "device_name": "Intel Arc Pro B60 24GB",
      "vendor": "Intel",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 456,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-pro-b65-32gb",
      "device_name": "Intel Arc Pro B65",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "intel-arc-pro-b70-32gb",
      "device_name": "Intel Arc Pro B70",
      "vendor": "Intel",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-a100-80gb-pcie",
      "device_name": "NVIDIA A100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 1935,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-a100-80gb-sxm",
      "device_name": "NVIDIA A100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2039,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-b200",
      "device_name": "NVIDIA B200",
      "vendor": "NVIDIA",
      "device_memory_gb": 180,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-b300",
      "device_name": "NVIDIA B300",
      "vendor": "NVIDIA",
      "device_memory_gb": 270,
      "device_bandwidth_gb_s": 7700,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-dgx-spark",
      "device_name": "NVIDIA DGX Spark",
      "vendor": "NVIDIA",
      "device_memory_gb": 128,
      "device_bandwidth_gb_s": 273,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1060-6gb",
      "device_name": "NVIDIA GeForce GTX 1060 6GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.2,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1070",
      "device_name": "NVIDIA GeForce GTX 1070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1080",
      "device_name": "NVIDIA GeForce GTX 1080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320.3,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1080-ti",
      "device_name": "NVIDIA GeForce GTX 1080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 484.4,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1650",
      "device_name": "NVIDIA GeForce GTX 1650",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 128.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1660",
      "device_name": "NVIDIA GeForce GTX 1660",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192.1,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1660-super",
      "device_name": "NVIDIA GeForce GTX 1660 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-gtx-1660-ti",
      "device_name": "NVIDIA GeForce GTX 1660 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2060",
      "device_name": "NVIDIA GeForce RTX 2060",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2060-12gb",
      "device_name": "NVIDIA GeForce RTX 2060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2060-super",
      "device_name": "NVIDIA GeForce RTX 2060 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2070",
      "device_name": "NVIDIA GeForce RTX 2070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2070-super",
      "device_name": "NVIDIA GeForce RTX 2070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2080",
      "device_name": "NVIDIA GeForce RTX 2080",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2080-super",
      "device_name": "NVIDIA GeForce RTX 2080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 495.9,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-2080-ti",
      "device_name": "NVIDIA GeForce RTX 2080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 11,
      "device_bandwidth_gb_s": 616,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3050-8gb",
      "device_name": "NVIDIA GeForce RTX 3050 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 224,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3050-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3050 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 4,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3060-12gb",
      "device_name": "NVIDIA GeForce RTX 3060 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 360,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3060-laptop",
      "device_name": "NVIDIA GeForce RTX 3060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 336,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3060-ti",
      "device_name": "NVIDIA GeForce RTX 3060 Ti (GDDR6)",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3070",
      "device_name": "NVIDIA GeForce RTX 3070",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3070-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3070-ti",
      "device_name": "NVIDIA GeForce RTX 3070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 608,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3080",
      "device_name": "NVIDIA GeForce RTX 3080",
      "vendor": "NVIDIA",
      "device_memory_gb": 10,
      "device_bandwidth_gb_s": 760,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3080-12gb",
      "device_name": "NVIDIA GeForce RTX 3080 12GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3080-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3080-ti",
      "device_name": "NVIDIA GeForce RTX 3080 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 912,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3080-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 3080 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 512,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3090",
      "device_name": "NVIDIA GeForce RTX 3090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 936,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-3090-ti",
      "device_name": "NVIDIA GeForce RTX 3090 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4050-laptop",
      "device_name": "NVIDIA GeForce RTX 4050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 6,
      "device_bandwidth_gb_s": 192,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4060",
      "device_name": "NVIDIA GeForce RTX 4060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 272,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4060-laptop",
      "device_name": "NVIDIA GeForce RTX 4060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 4060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 288,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4070",
      "device_name": "NVIDIA GeForce RTX 4070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4070-laptop",
      "device_name": "NVIDIA GeForce RTX 4070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 256,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4070-super",
      "device_name": "NVIDIA GeForce RTX 4070 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4070-ti",
      "device_name": "NVIDIA GeForce RTX 4070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 504,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4070-ti-super",
      "device_name": "NVIDIA GeForce RTX 4070 Ti SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4080",
      "device_name": "NVIDIA GeForce RTX 4080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 716.8,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4080-laptop",
      "device_name": "NVIDIA GeForce RTX 4080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 432,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4080-super",
      "device_name": "NVIDIA GeForce RTX 4080 SUPER",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 736,
      "bandwidth_evidence": "derived",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4090",
      "device_name": "NVIDIA GeForce RTX 4090",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 1008,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-4090-laptop",
      "device_name": "NVIDIA GeForce RTX 4090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 576,
      "bandwidth_evidence": "founded",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5050",
      "device_name": "NVIDIA GeForce RTX 5050",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 320,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5050-laptop",
      "device_name": "NVIDIA GeForce RTX 5050 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5060",
      "device_name": "NVIDIA GeForce RTX 5060",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5060-laptop",
      "device_name": "NVIDIA GeForce RTX 5060 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5060-ti-16gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 16GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5060-ti-8gb",
      "device_name": "NVIDIA GeForce RTX 5060 Ti 8GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 448,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5070",
      "device_name": "NVIDIA GeForce RTX 5070",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5070-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 8,
      "device_bandwidth_gb_s": 384,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5070-ti",
      "device_name": "NVIDIA GeForce RTX 5070 Ti",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5070-ti-laptop",
      "device_name": "NVIDIA GeForce RTX 5070 Ti Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 12,
      "device_bandwidth_gb_s": 672,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5080",
      "device_name": "NVIDIA GeForce RTX 5080",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5080-laptop",
      "device_name": "NVIDIA GeForce RTX 5080 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 16,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5090",
      "device_name": "NVIDIA GeForce RTX 5090",
      "vendor": "NVIDIA",
      "device_memory_gb": 32,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-geforce-rtx-5090-laptop",
      "device_name": "NVIDIA GeForce RTX 5090 Laptop GPU",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 896,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-h100-80gb-pcie",
      "device_name": "NVIDIA H100 80GB PCIe",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 2000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-h100-80gb-sxm",
      "device_name": "NVIDIA H100 80GB SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 80,
      "device_bandwidth_gb_s": 3350,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-h200-nvl",
      "device_name": "NVIDIA H200 NVL",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-h200-sxm",
      "device_name": "NVIDIA H200 SXM",
      "vendor": "NVIDIA",
      "device_memory_gb": 141,
      "device_bandwidth_gb_s": 4800,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-l4",
      "device_name": "NVIDIA L4",
      "vendor": "NVIDIA",
      "device_memory_gb": 24,
      "device_bandwidth_gb_s": 300,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-l40s",
      "device_name": "NVIDIA L40S",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 864,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rtx-6000-ada",
      "device_name": "NVIDIA RTX 6000 Ada Generation",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 960,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rtx-a6000",
      "device_name": "NVIDIA RTX A6000",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 768,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-48gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 48GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 48,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rtx-pro-5000-blackwell-72gb",
      "device_name": "NVIDIA RTX PRO 5000 Blackwell 72GB",
      "vendor": "NVIDIA",
      "device_memory_gb": 72,
      "device_bandwidth_gb_s": 1344,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rtx-pro-6000-blackwell-workstation",
      "device_name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition",
      "vendor": "NVIDIA",
      "device_memory_gb": 96,
      "device_bandwidth_gb_s": 1792,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    },
    {
      "model_slug": "zyphra-zamba2-1-2b-instruct",
      "model_name": "Zamba2-1.2B-instruct",
      "publisher": "Zyphra",
      "hf_repo": "Zyphra/Zamba2-1.2B-instruct",
      "config_revision": "c06b789753995762b3c11b3eeb39e38634f897e9",
      "parameters": 1215064704,
      "device_id": "nvidia-rubin",
      "device_name": "NVIDIA Rubin",
      "vendor": "NVIDIA",
      "device_memory_gb": 288,
      "device_bandwidth_gb_s": 22000,
      "bandwidth_evidence": "verified",
      "context_tokens": 8192,
      "status": "context_exceeds_model_max",
      "status_reason": "The model's published maximum is 4096 tokens, below the 8192-token reference context of this dataset.",
      "fits": null,
      "best_format": null,
      "required_bytes": null,
      "utilisation": null,
      "weight_size_evidence": null,
      "smallest_format": null,
      "shortfall_bytes": null
    }
  ]
}